¹ Musch, H., Costanzo Festa als Madrigalkomponist, Sammlung musikwissenschaftlicher Abhandlungen 61 (Baden-Baden, 1977), pp. 157–69.Google Scholar Additional pieces and some new concordances are cited in Fenlon, I. and Haar, J., ‘Fonti e cronologia dei madrigali di Costanzo Festa’, Rivista Italiana di Musicologia, 13 (1978), on pp. 231–42.Google Scholar

² For a study of this volume and its multitudinous editions, see Bridges, T. W., ‘The Publishing of Arcadelt's First Book of Madrigals’ (Ph.D. dissertation, Harvard University, 1982).Google Scholar A modern transcription, with comments on the earlier editions, appears in Jacobi Arcadelt opera omnia, ed. Seay, A., Corpus Mensurabilis Musicae 31/ii (n.p., 1970).Google Scholar This edition will be cited hereinafter as Arcadelt.

For many of the printed volumes mentioned or discussed in this paper, the references and descriptions in RISM (Schlager, K., ed., Einzeldrücke vor 1800, Répertoire International des Sources Musicales A/I, Kassel, 1971–1981Google Scholar [designated by letter and number]; F. Lesure, ed., Recueils imprimés, XVIe–xviie siècles, ibid. B/iv/1, Munich and Duisburg, 1960 [designated by date with superscript number]) or the New Vogel (Lesure, F. and Sartori, C., eds., Bibliografia della musica italiana vocale profane pubblicata dal 1500 a1 1700, Pomezia, 1977;Google Scholar rev. edn of Vogel, E., Bibliothek der gedruckten weltlichen Vocalmusik Italiens, aus den Jahren 1500 bis 1700, Berlin, 1892)Google Scholar are not satisfactory or sufficient to enable the reader to distinguish editions or single copies. Therefore, at the first reference of substance to a new volume, I give below a transcription of the title-page, together with such notes as are necessary to supplement these reference works. In my transcriptions and notes, which are not intended to be comprehensive, the word ‘flower’ refers to a printer's flower, which need not represent a flower as such; the collation, when given, arranges the partbooks in the standard order of Cantus, Tenor, Altus, Bassus, Quintus, etc. Thereafter the volume is cited by its reference siglum in RISM (whenever possible) in order to save excessive duplication of bibliographical detail.

References to composers' names appearing in the early editions are here given in italics if they are direct quotations of the original form. In such cases, a final point may well appear (if part of the original), and this should not be confused with the end of a sentence in the present text.

³ These are discussed by Chapman, C. W. in her doctoral dissertation, ‘Andrea Antico’ (Harvard University, 1964).Google Scholar

⁴ This is something that is not always found once single-impression type takes over from blocks. With wood-blocks, there was a strong incentive for starting new pieces at the beginnings of new lines, for this gave the printer the flexibility to alter the order of items in subsequent editions, in exactly the manner permitted by blocks prepared for a volume of drawings or of maps. There is no reason to suppose that the Antico–Scotto partnership did not anticipate the possibility of later editions: the history of earlier music printers – Petrucci, Antico himself, or the Dorico brothers – suggests that this was seen as a normal practice, while some of the first Scotto editions were themselves taken from Antico blocks previously printed by other men.

⁵ RISM V1223; New Vogel 2887. Both suggest the date ca[1535].

The Altus and Bassus partbooks survive, labelled

Madrigali a cinque Libro primo. /A [B].

Oblong quarto-in-eights: EF⁸; GH⁸.

The Tavola, on the first verso, is in one column, in order of the compositions in the book, and is of the same setting of type for both parts. This setting was probably used for all four parts, for it lists the nineteenth piece as Deh non gionger tormenti; these words are the second phrase of the madrigal Purtroppo donn'in van', and the first words sung by the Cantus, Altus and Quintus parts. This piece is misnumbered 20 in the Bassus, fol. H6^r; Altro non e'l mio amor, the twelfth piece, is misnumbered 11 in the Altus.

The copies consulted for this study were, Altus: Paris, Bibliothèque Nationale, Rés. Vmd. 30; Bassus: Bologna, Civico Museo Bibliografico Musicale, R 140(2); both complete.

For the actual date of this volume see below, note 62. Pace the New Vogel, these partbooks contain sixteen folios (i.e. thirty-two pages) each, as probably did the other partbooks; the last folio of the Bassus is completely blank.

⁶ RISM 1541¹⁷; New Vogel 2885.

[flower] CANTVS [flower] / LE DOTTE, ET ECCELLENTE COMPOSITIONI DE I MA=/drigali di VERDELOT, A cinque Voci, & da diversi perfettissimi Musici fat=/te. Novamente ristampate, & con ogni diligentia correte. / m.d. [Gardane's mark] xxxxi. / Excudebat Venetiis, apud Antonium Gardane.

The title-page also carries the signature line: Primi, Verdelot a cinque. A

The other voice parts carry the same title-page (with changed part-name) and signature line (with changed letter).

Oblong quarto, 48 pages per part: A–F⁴; G–M⁴; N–S⁴; T–Z.AA⁴; BB–GG⁴.

The copy consulted is at Washington, D.C., Library of Congress. The Cantus partbook at Paris, Bibliothèque Nationale, Rés. 1169 is of the same press-run.

⁷ This is one volume where the pieces do not always open at the start of a line: the need to fit Barre's Come potro fidarmi and Baldassare d'Imola's Non vi gloriate onto a single opening, as well as that of squeezing Verdelot's Donne se fiera Stella after Barre's I sospiri amorosi, led to the start of each second piece appearing part-way along a line in one or two of the partbooks.

⁸ Examples would include fol. Y3‘ (p. 45) of the Altus of RISM 1541¹⁸, or K1^v (p. 26) of the Tenor of 1541¹¹.

⁹ RISM 1541¹⁸ = V1229; New Vogel 2872.

[flower] CANTVS [flower] / DI VERDELOT TVTTI LI MADRIGALI DEL PRIMO ET SE=/condo Libro, a Quatro Voci: Novamente ristampati, & da molti errori emen/dati. Con la gionta de i Madrigali del medesmo Autore. / AGGIONTOVI ANCHORA ALTRI MADRIGALI / novamente Composti da Messer Adriano, & da Altri Eccellentissimi Musici. / md [Gardane's mark] xxxxi / Excudebat Venetiis, apud Antonium Gardane.

The lower voices have the same title (with the part-name changed) and with the addition of a signature line: Verdelot primi, & secundi. [with the appropriate signature letter]

Oblong quarto, 64 pages per volume: A–H⁴; I–Q⁴; R–Z, AA⁴; BB–II⁴.

The copy consulted is at Vienna, Österreichische Nationalbibliothek, sa.77.d.53.

Typically, the composer listings in the entry in New Vogel are far from consistent. Of the pieces which straddle an opening, and which have no ascription of their own, Non vi fidat'o simplicett'amanti (pp. 6–7) is assigned to the composer on the verso (almost certainly incorrectly, as I am about to argue), while Io son tal volta (pp. 50–1) is left anonymous. Igno soave lies across two pages that both carry the name of Verdelot, and should perhaps be assigned to him in any account of this edition, while Vostre dolce parole (pp. 16–17) has the name Iachet fitted above its first stave in three of the four partbooks.

¹⁰ These three have survived elsewhere, without conflicting attributions, as the work of Verdelot, Jachet and Verdelot, respectively. The present discussion does not throw any doubt on this pattern, but is offered solely to draw attention to the procedure adopted by the printer.

¹¹ It has to be admitted that there is no firm bibliographical evidence here for the authorship of the second piece on pp. 6–7. My contention, supported by the concordance pattern and by other evidence (to be advanced), is that a piece in this position is normally, by this time, to be assigned to the composer named at the head of the verso, although by no means consistently following that pattern. Indeed, such a view raises questions about the opinion which Gardane held of the authorship of Io son tal volta (G1^v–2^r): however, as I shall show, this kind of ambiguous situation is just that which led to changes of ascription as printer's habits became more settled and consistent. I shall return to this, and some other problems of ascription in the present edition, in connection with a different range of evidence.

¹² RISM 1541¹¹ = A1376;New Vogel 158.

[flower] CANTVS [flower] / ILTERZO LIBRO DE I MADRIGALI NOVISSIMI/di archadelt a quattro voci insieme con alchuni di constantio festa & altri diecibellissimi / a voci mudate novamente ristampati con nova gionta & nova corretione. / m.d. [Gardane's mark] l.xi. / Venetiis apud antonium gardane

The other partbooks have the same title (with changed voice-name), plus a signature line: Terzo Libro d'archadelt. [plus signature letter]

Oblong quarto, 48 pages per volume: A–F⁴; G–M⁴; N–S⁴; T–Z, + ⁴.

The copy consulted is at London, British Library, K.2.h.5. The second part of the date, l.xi., is changed in manuscript to xli.

Again, New Vogel is in error, or misleading, in its ascriptions for this volume: E morta la speranza, spread between pp. 12 and 13, is strictly anonymous– the caption Archadelt. heads p. 12, while Con. festa. appears on p. 13 (this might be another case for arguing that the piece starting at the foot of the verso was thought to be by the composer named at the head of that page, for this madrigal is ascribed to Arcadelt in 1539²³); Se i sguardi di costei is only attributed to Con. festa in the Cantus and Tenor books; Lasso chepur hormai, on pp. 26–7, is anonymous; Madonna s'io credessi (p. 29) is clearly attributed to Archadelt.; Languir non mi fa amore is only ascribed to Corteccia. in the lower three voices, while the Cantus reads Archadelt. (see below); and S'altrui d'amor is clearly ascribed to Archadelt. in all voices. (Some of these works will be discussed below. One must deplore the haphazard procedure adopted in New Vògel (for this and other volumes) whereby the name of the composer to whom the whole volume is dedicated is not repeated against individual madrigals: it does not allow the reader to distinguish between anonymous works and those that appear below the name of the composer given on the title-page, in this case, Arcadelt.)

In stating that 1541¹¹ is the second edition of this title, I am making the following observations: (1) that RISM [c1556]²³, copy at Venice, Biblioteca Nazionale Marciana, Mus. 365–7 (Tenor, Altus and Bassus) is of the same edition as 1539²³ = A1374 (Tenor, Altus and Bassus of the copy at Munich, Bayerische Staatsbibliothek, 4°. Mus.pr. 95/2), as has been noted by other scholars (see Arcadelt, iv, p. ix, or New Vogel, no. 156), and that this is probably the first edition; (2) that the Cantus of the copy at Munich represents part of the same edition, although cited separately in New Vogel as no. 157; (3) that the remaining editions are as cited in RISM: 1541¹¹ =A1376; 1543²⁰ = A1377; 1556²² = A1378.

The bibliographical evidence for asserting that the Cantus book at Munich is part of the same edition as the other partbooks is far from clear. The most difficult problem is a direct result of the Cantus carrying a dedication, on fol. A1^v: although the music follows the same sequence as that found in the other partbooks, it cannot be arranged in the same manner, for it starts on a recto, rather than on a verso. Where the Cantus places three pieces on an opening, it cannot reflect exactly the layout of the other three parts, for there these pieces would occupy a recto followed by its own verso, and would involve a page-turn in the middle of the second madrigal. The solution adopted was to use the same openings for three pieces, which results in these being different works: the following extract from a table of contents will make this clear, indicating the folio and (in roman numerals) the stave on which the pieces start in each book:

This situation is most unusual, and would argue, at first sight, for the Cantus being part of a separate edition, potentially earlier than the others, despite the comment on the title-page that the pieces are ‘corretti’. Perhaps supporting such a conclusion could be the different wordings found on the title-pages:

RISM 1539²³ = A1374; New Vogel 156–7.

[Cantus:] IL TERZO LIBRO DE I MADRIGALI / NOVISSIMI DI ARCHADELTH A QVATTRO VOCI, / Insieme con alchuni di Constantio Festa, & altri died bellissimi a Voci mudate. / Novamente con ogni diligentia Stampati, & corretti. / [flower] / LIBRO TERZO [Scotto's mark] A QVATTRO VOCI. / VENETIIS / APVD HIERONYMVM SCOTVM. / [short rule] / 1539.

The other voices read:

DEL TERTIO LIBRO DE I MADRIGALI / DI ARCHADELT, ET DI ALTRI ECCELLENTISSIMI / Authori. Con la gionta de alcuni Madrigali a Voci mutate bellissimi / A QVATTRO VOCI. / T / LIBRO TERTIO or, for the Altus and Bassus:

DEL TERZO LIBRO […] A [B] / LIBRO TERZO

The presence of these differing title-pages is not a very strong argument for their belonging to different editions, for a number of volumes printed by the Scottos at this period carry different title-pages for the lower voices.

However, there are three other features, each of which specifically leads me to believe that all four parts are probably of the same edition. One is the apparently random use of the word Tertio as opposed to Terzo, and that of Arcadelth as opposed to Archadelt, in the signature line for each gathering. Tertio is used only rarely, on fols. a1^r and 1^r and 2^r on gatherings b and c, as well as in the title of the Tenor book: Terzo appears everywhere else. Arcadelth is found more frequently: B1^r of the Cantus; b1^r, c2^r, d1^r, d2^r, e1^r, e2^r, f1^r and f2^r of the Tenor; throughout the Altus; and on ff1^r of the Bassus book. (Note that the Venice copy has the same pattern as that found in the Munich books.)

Second is the state of the initial letters used in these partbooks. Those in the first three gatherings of the Tenor are marginally better than elsewhere. On the other hand, those in the Cantus are neither better nor worse overall than in the other books, tending merely to conform with the worst existing state elsewhere.

The sum of these two pieces of evidence argues that the first three gatherings of the Tenor were the first to be prepared and printed. The rest of the Tenor then made one complete process, with the Altus and the Bassus: the Cantus was either part of the same process, or followed it immediately, without the intervention of any other printing.

But the decisive factor is that of the TAVOLA, appearing at the end of each book: this shows quite clearly that all four copies were printed from the same setting of type, and that the Cantus was printed last. There are idiosyncratic spacings and alignments repeated from book to book, as well as one or two places that have taken ink in identical manner. Apart from the normal and inevitable changes demonstrating that the Cantus was last in the sequence, there is a further detail: the tavola has, in addition to the normal two columns of contents, two additional columns flanking them, which contain the words A voce pari., when needed. In the Cantus alone, the left column has slipped down half a line.

The apparent conclusion has to be that the Cantus was printed alone, and immediately after the other books, before the type for the tavola was dispersed. The probable reason for this is to be found in the presence of the dedicatory letter printed on fol. A1^v, This letter, though signed Geronimo Scotto, draws attention to the desire of his brother, Ottaviano Scotto, to compliment the dedicatee, the Rev. Mons. Girolamo Verallo, then Papal Legate in Venice. Verallo was translated to the See of Bertinoro in February of 1539/40, which is sufficient to confirm that the date of 1539 on the title-page of the Cantus book is a genuine one, not falsified to conform with the other books. Gerolamo Scotto took over the printing shop from Ottaviano during 1539, and it is quite plausible that this dedication, with the consequent rearrangement of the Cantus book, represents one of his first publications as an independent printer.

¹³ For this reason, I should expect 1541¹⁸ to have been prepared for the press rather later in the year than was the present volume. The arrangement of the composers' names is more in line with later practice, although not completely straightforward. However, it is worth bearing in mind Mary Lewis's remarks about the extent to which Gardane seems to have worked on several volumes simultaneously: see her ‘Antonio Gardane and his Publications of Sacred Music, 1538–1555’ (Ph.D. dissertation, Brandeis University, 1979), pp. 139–58.

¹⁴ This edition is not discussed in the present study. To demonstrate that the conventions outlined here continued in use for another fifteen years would require a considerable extension to this paper, an extension of no great relevance and, further, one which I am not yet in a position to offer.

¹⁵ Despite the suggestion, in Arcadelt, iv, p. xii, this is not strictly anonymous in 1543²⁰, for the same name, Archadelt, is found at the head of both relevant pages, xx and xxi. By 1543, Gardane was regularly implying that the composer of the piece straddling the opening was the name he printed at the head of the verso, unless otherwise indicated.

¹⁶ Musch, Costanzo Festa, pp. 158–67, and Fenlon and Haar, ‘Fonti’, pp. 237–8, both list the works, while Seay's edition of Arcadelt prints all three. It may be that all should be reconsidered, in the light of the types of evidence to be offered in section iii of this paper, below.

¹⁷ Lest it be thought that the ascription in 1539²³ could be a simple error (as is, for example, that to Madonna s'io credessi; cf. Haar, J. and Bernstein, L., eds., lhan Gero: Il primo libro de' madrigali italiani et canzoni francese a due voci, Masters and Monuments of the Renaissance 1 (New York, 1980))Google Scholar, it is worth remarking that the pieces ascribed to Festa are all to be found on fols. B3^v–C3^r, eight pages which form a discrete unit (and which, incidentally, may suggest something about how the music for this volume was collected). It should be added that both Musch and Seay (in his edition of Arcadelt) treat the ascriptions as if they were handled inconsistently by Gardane, in each case thereby supplying reinforcement of the author‘s position.

¹⁸ For this madrigal, see below.

¹⁹ In the first edition, this madrigal, Lasso chepur hormai, falls comfortably within the group of works assigned to Festa, while in the second there are no bibliographical grounds for rejecting the printed ascription to the same composer.

²⁰ The wording of this phrase is in accordance with a point which I shall be pursuing in section iii of the present study, to the effect that apparent ascriptior are sometimes no more than the use of a name to act as a running head-line, that is, as an internal title.

²¹ They are Amor s'al primo sguardo and Si lieto alcun giammai, both to be discussed in section ii below, and Divelt'el mio bel vivo, Languir non mi fa amore and Poi che'l fiero destin, all of which will be covered in section iii.

²² Musch, Costanzo Festa, p. 166.

²³ First edition: RISM 1539²⁴ = A1379; New Vogel 161.

IL QVARTO LIBRO DI MADRIGALI D'ARCHADELT A / QVATRO VOCI COMPOSTI VLTIMAMENTE INSIEME CON / ALCVNI MADRIGALI DE ALTRI AVTORI NOVAMENTE CON / OGNI DILIGENTIA STAMPATI ET CORRETTI. / CANTVS [mark] CANTVS. / CON GRATIA ET PRIVILEGIO.

The other voices carry the same title, with the part-name changed. The word Tenor is misspelled TENNR to the right of the mark.

All parts carry a colophon on the last verso:

IN VENETIA NELLA STAMPA D'ANTONIO GARDANE / Nellanno del Signore m.d. xxxix. Nel mese di Setembre. / [mark] / CON GRATIA ET PRIVILEGIO.

Oblong quarto, 40 pages per part: A–E⁴; F–K⁴; L–P⁴; Q–V⁴.

Copy consulted: Munich, Bayerische Staatsbibliothek, 4°. Mus.pr. 95/3.

Second edition: RISM 1541¹² = A1380; New Vogel 162.

[flower] CANTVS [flower] / IL QVARTO LIBRO DI MADRIGALI D'ARCHA=/delt, a Qvattro Voci, Composti vltimamente insieme con alcvni Madrigali d'altri aut = /tori, Novamente con ogni diligentia ristampati, & corretti. / m. d. [Gardane's mark] xxxxi. / NON SINE PRIVILEGIO. / Excudebat Venetiis, apud Antonium Gardane.

The lower voices have the same title, with the change of the part-name and the addition of a signature line:

Quarto libro d’Archadelt. [plus signature letter]

Oblong quarto, 40 pages per part: A–E⁴; F–K⁴; L–P⁴; Q–V⁴.

Copy consulted: London, British Library, K.2.h.6.

Third edition: RISM 1545¹⁸ = A1381; New Vogel 163.

[flower] ARCHADELT [flower] / QVARTO LIBRO / DI MADRIGALI A QVATRO VOCI D'ARCHADELT / Insieme alcuni di altri autori novamente ristampato et corretto / A QVATRO [Gardane's mark] VOCI / Venetijs Apud Antonium Gardane / [rule] / m. d. xxxxv. / CANTVS

The lower voices have the same title, with changed part-name.

Oblong quarto, 32 pages per part: A–D⁴; E–H⁴; I–M⁴; N–Q⁴. Paginated in roman numerals from the second recto, i–xxix, [xxx].

Copy consulted: Bologna, Civico Museo Bibliografico Musicale, r 58.

New Vogel again makes a number of errors in ascriptions for these three editions: Dal bel suave ragio is not ascribed to Layolle in the first edition – the word Archadelt. appears at the head of the relevant page (see below); Col pensier mai is anonymous in the first edition, as (strictly) is S'era forsi ripreso, for the latter appears at the foot of a page headed with Arcadelt's name; Pace non trovo is only ascribed to Yvo in the second edition; finally, New Vogel fails to point out that eight pieces (nos. 10, 14, 18, 23, 24, 26, 27 and 29) are anonymous in the third edition.

²⁴ There are two elements to this argument. First is the evidence that two typesetters were involved in setting the head-lines, and probably all the text. One man set the outer forme of the first gathering of each book (A, E, I and N), and the other did the inner. If we call these men X and Y, then X set 1^r, 2^V, 3^r and 4^V, while Y was responsible for 1^v, 2 ^r;, 3^V and 4^r. The second gathering (B, F, K and O) has all the captions re-set: X worked on the inner forme and Y on the outer. In the third gathering (C, G, L and P), as many captions as possible were retained: it appears that Y set both formes. The fourth gathering (D, etc.) has newcaptions: it is likely that Y set both formes again, although only two pieces have composers’ names attached to them. Neither of these can have been retained from the previous gatherings. Col pensier, on D3^r, in particular, does not retain the head-line Archadelt, found on C3^r.

Since there are no ascriptions on the outer forme of the last gathering (1^r, 2^V, 3^r and 4^V of D, H, M and Q–4^V contains the TAVOLA Delli Madrigali), one could suggest that all were omitted in error. However, the typesetter had another element to consider: five of the pieces in this gathering also have in the head-line the phrase A voce Pari. These five include two where the phrase does follow an ascription to Archadelt, and three (1^r, 1^v and 4^r) where the work is anonymous. In these three cases, the phrase is placed in exactly the space that would otherwise have been occupied by a composer's name. It seems likely, therefore, that the typesetter did not simply forget to insert an ascription, but rather had to determine whether an ascription, or the new phrase, or both, were needed for each page in this gathering. As a result, it would seem that the omissions of composers' names were deliberate.

²⁵ RISM 1545¹⁹ = V1231; New Vogel 2874.

The first folio of the Cantus part is missing in the only surviving copy. The lower voices have identical titles, with the exception of the part-name:

TENOR / VERDELOT TVTTI LI MADRIGALI DEL PRIMO / ET SECONDO LIBRO A QVATRO VOCI WOVAMENTE RI = /STAMPATI, ET CON DILIGENTIA CORRETTI. / A QVATRO VOCI / VENETIIS m. d. xlv. Oblong quarto, 40 pages per part: A–F⁴; G–M⁴; N–S⁴; T–Z. + ⁴.

Copy consulted: Florence, Biblioteca Nazionale Centrale, Mus.ant. 129: lacks fol. Al.

Mary Lewis discusses this and a number of other unsigned printed volumes of the period (‘Antonio Gardane’, pp. 314–29). She shows clearly that the typographical material for this volume was that normally used by Scotto, and assigns the book to him as publisher. However, the different treatment of certain details leads me to think that the book was edited (and possibly even set) by someone other than Scotto's normal house-men. Bridges, ‘Publishing’, pp. 119–29, reaches a similar conclusion, though for different reasons.

It may be significant that two of Lewis's, ‘Group 2’ (to which the present volume belongs) are probably among those volumes that, as Richard Agee has recently shown, were supported by privileges issued in Venice – see his ‘The Venetian Privilege and MusicPrinting in the Sixteenth Century’, Early Music History, 3 (1983), pp. 1–42Google Scholar: on pp. [29–30] are references to privileges issued to Rore for motets (22 November 1544, and therefore plausibly to be sought in the publications of 1545) and to Cambio for madrigals of Petrarch (2 June 1545). Others of these unsigned volumes, those by Festa, by Vicentino and of Madrigali de diversi autori (RISM 1547¹³) claim the existence of privilege. It is notable that, in the same period, Gardane received a privilege for the works of Jacques da Ponte, while in September 1544 Scotto gained a privilege for several different volumes. None of these has survived among the unsigned volumes that Lewis discusses: but they do indicate that both Gardane and Scotto were able and willing to apply for privileges on their own behalf during this period. It is not an unreasonable assumption that neither was involved in the petitions for privileges for the unsigned volumes.

However, both Rore and Cambio are among the musicians referred to by Andrea Calmo at just this period (1547) as being among the leaders of the period, alongside Arcadelt, Verdelot and Willaert. (The relevant extract from the later, 1580, edition of Calmo's letters is quoted and translated in Bridges, ‘Publishing’, pp. 57–8.) While neither Verdelot nor Arcadelt is known to have been in Venice at the time, all five composers are represented among the unsigned volumes discussed by Lewis, including Verdelot with the volume which stimulated this digression. It seems likely that a number of the unsigned volumes were published, or at least covered by privilege, by the composers concerned (as is suggested by the evidence of the privileges to Cambio and Rore), and merely printed by Scotto. This would certainly help to explain why no printer's name appears on the volumes, even after Scotto began inserting the salamander device on the title-pages. It would also raise interesting speculation as to where Verdelot was living during the 1530s and 1540s. If he was in Venice, this would to some extent help to remove the peculiar situation of a large number of volumes of music being published in Venice during the 1530s as the work of a non-resident. There is no parallel case during the period, if Arcadelt was resident, for he was not represented until the end of the decade (by which time the situation was changing rapidly): Festa had very few volumes to show, and, further, himself took out a privilege in 1538, as Agee shows (in ‘Privilege’, p. [29]). See Appendix, below.

²⁶ RISM 1540²⁰ V1228; New Vogel 2871.

[flower] DI VERDELOTTO [flower] / TVTTI LI MADRIGALI DEL PRIMO, ET SECONDO / Libro a Qvatro Voci. Con la Gionta de i Madrigali del medesmo / Auttore, non piu stampati. / AGGIONTOVI ANCHORA ALTRI MADRIGALI / novamente Composti da Messer ADRIANO, & de altri Eccellentissimi / Musici, Come appare ne la sequente Tavola. / [flower] / [device] / Apud Hieronymum Scotum. / [rule]/1540.

The lower voices have a different title, with the relevant part-name:

MADRIGALI DEL PRIMO, ET SECONDO LIBRO / di Verdelotto a Quatro Voci. Con la Gionta del medesmo Auttore, & de / altri Eccellentissimi Musici, novamente Stampati. / [flower] / TENOR

Oblong quarto, 60 pages per part: A–G⁴H²; a–g⁴h²; AA–GG⁴HH²; aa–gg⁴hh². Paginated from the second recto, i–lviii.

Copy consulted: Wolfenbüttel, Herzog August Bibliothek, 2.13.14–2.13.17. Musica.

New Vogel does not correctly reflect the fact that eleven of the pieces lying across an opening are not specifically attributed, while the others are. Further, it suggests that the second setting of Con lachrim'et sospir (pp. xliv–xlv, fols. F3^v–4^r of each book) is by Verdelot: the names that appear at the heads of the relevant pages are ARCHADELT and IACHET, respectively. Finally, Amor quanta piu lieto (p. xliii) is clearly ascribed to VERDELOT in all books. Both these pieces are discussed below.

²⁷ Citations appear below, at note 45. For later sources, see the list in Musch, Costanzo Festa, p. 161, as supplemented by Fenlon and Haar, ‘Fonti’, p. 232.

²⁸ For a discussion of this in the output of a major early Italian printer, see my ‘Upon the Use of Running Titles in the Aldus House of 1518’, The Library, 5th ser., 27 (1972), pp. 126–31.Google Scholar I use the same evidence to help in the analysis of Petrucci's printing methods, in the forthcoming paper, ‘A Case of Half-sheet Imposition in the Early Sixteenth Century’.

²⁹ Lewis, ‘Antonio Gardane’, pp. 117–18, has suggested that Gardane may have employed distinctively shaped formes in order to keep the staves in the same relationships from page to page. Some such system would have been essential for Petrucci and other multiple impression printers.

³⁰ For more discussion of the implications of this procedure, see section iii of the present paper, below.

³¹ RISM 1540¹⁹ = VI087; New Vogel 2844.

[flower] MADRIGALI [flower] / A QVATTRO VOCI / Di Messer CLAVDIO Veggio. con la Gionta di sei altri di Arcadelth della / misura a breve. Nuovamente con ogni diligentia stampati. / [device] / VEXETIIS / Apud Hieronvmum Scotum. / [rule] / 1540.

The lower voices have a different title-page:

MADRIGALI DI MESSER CLAVDIO / Veggio, a Quattro Voci. Nuovamente Stampati. / [flower] / TEXOR / [etc.] Oblong quarto, 40 pages per part: A–E⁴; a–e⁴; AA–EE⁴: aa–ee⁴. Paginated from the second recto, i–xxxviii.

Copy consulted: Vienna, Österreichische Nationalbibliothek, SA.77.D.55.

Haar, J., ‘The Nota Nere Madrigal’, Journal of the American Musicological Society, 18 (1965), pp. 22–41CrossRefGoogle Scholar, points out (on p. 25) that this is the first printed volume to refer to the new style.

³² Lewis, ‘Antonio Gardane’, pp. 123–5. The evidence of the treatment of the head-lines in Veggio's book, which I have just outlined, argues that Scotto was following the same procedure.

³³ RISM 1541¹⁶; New Vogel 2890.

VERDELOT / LA PIV DIVINA, ET PIV BELLA MVSICA, CHE SE / vdisse giamai delli presenti Madrigali, a Sei voci. Composti per lo Eccellentissi=/mi VERDELOT. Et altri Musici, non piu Stampati, & con / ogni diligentia corretti. Novamente posti in luce. / CANTVS / CON GRATIA ET PRIVILEGIO. / m.d.xli. / VENETIIS APVD ANTONIVM GARDANE.

The lower voices have different titles, with the appropriate part-name:

TENOR / VERDELOT / m. d. [Gardane's mark] xxxxi. / Venetijs Apud Antonium Gardane. [plus signature line:] Madrigali primi, di Verdelot, a Sei [followed by the signature letter]

Oblong quarto, 32 pages per part: A–D⁴; E–H⁴; I–M⁴; N–Q⁴; R–V⁴; X–Z,+⁴. Paginated from the first recto, [1], 2–31, [32] in all books.

Copies consulted: Vienna, Osterreichische Nationalbibliothek, SA.77.D.52; London, British Library. K.ll.e.2(4) – Bassus only.

Once again, there are serious problems with the list of ascriptions offered by the New Vogel: Ultimi mei sospiri is ascribed to Verdelot, as is Ardenti miei sospiri, while In me cresce l'ardore is anonymous. Four other pieces have ascriptions only in some of the partbooks, remaining anonymous in the others: all are discussed here.

³⁴ On the strength of this ascription, the work is edited in Arcadelt, vii, no. 27. The folios in the other partbooks of this edition correspond: G3^V, L3^V, P3^V, T3^V and Z3^V.

³⁵ This must serve to underscore the extent to which we should adapt our view of‘errors’, ‘variants’ and ‘changes’ in the surviving copies of early printed music– a view which is at present far from that of the contemporary purchaser of the volumes. For a printer to re-set a whole leaf or sheet, the errors had to be gross. For him to recall or correct pages that had slipped past him was virtually unknown. While Petrucci and other printers would regularly correct copies remaining unsold in their shop, supposing the errors to be serious musical ones, they felt almost no compulsion to correct non-musical mistakes. These were simply not deemed to be important enough, as is witnessed by the present erroneous ascription, which was merely ‘corrected’ for those partbooks which had not as yet gone through the press.

³⁶ Detailed analysis leads me to believe that the Tenor book was set first for the last gathering, which contains this madrigal: for earlier gatherings, the Cantus was certainly set first, perhaps followed by the Bassus.

³⁷ Despite the assertion of Bragard, A.-M., Étude bio-bibliographique sur Philippe Verdelot, musicien français de la Renaissance, Mémoires de la Classe des Beaux-Arts de l'Académie Royale de Belgique, 11 (1964), p. 58Google Scholar, this work is not to be found in later editions of Verdelot's sixvoice madrigals. It is discussed and edited in 1Harrán, D., ‘Chi bussa? or the Case of the Anti-madrigal’, Journal of the American Musicological Society, 21 (1968), pp. 85–93.CrossRefGoogle Scholar

³⁸ To be found on p. 6 – that is, fol. 3^V of the first gathering – of each partbook.

³⁹ This assumes that each gathering was set entirely through in consecutive order. One of the merits of the procedure proposed by Lewis is that the typesetter could not merely set the same gathering for each partbook before proceeding to the next, but he could also do the same for smaller units, certainly separate formes, and even possibly individual pieces. This last would in practice seem to have been unlikely, for, in the instance of the present volume, the music would have had to have been set for nearly six whole gatherings (one for each voice-part) before any one forme would have been completed and ready to go to press – to be precise, for thirty-seven pages of standing type. It is more likely that the procedure involved setting ‘vertically’ the individual pieces that made up one forme for each sheet. Then nineteen pages would have been set by the time the first forme was ready for the press. While this may seem still to make great demands on the amount of type, the number of sorts in the case, it must be remembered that one more page set would release another forme for the press, as that for the second partbook was finished, and another page would release the third, by which time the first would be returning from the press, and the type could be redistributed. Since fifteen pages of standing type would be required even under straightforward setting procedures (when each gathering was set straight through in order), this alternative arrangement does seem to have been feasible: indeed, it has the great advantage that each sheet would have time to dry from the first impression before the forme for the other side was printed, considerably greater time than that allowed by normal linear setting. However, this alternative method of vertical setting does weaken my argument for the attribution of Maistre Jhan‘s Madonna i prieghi mei, unless I assume that the lapse lies with the house editor or a shop supervisor, rather than with the actual typesetter.

⁴⁰ RISM 1546¹⁹; New Vogel 2891.

VERDELOT A SEI / MADRIGALI DI VERDELOT ET / DE ALTRI AVTORI A SEI VOCI / novamente con alcuni madrigali novi ristampati & corretto / A SEI [Gardane's mark] VOCI / In Venetia Apresso di / Antonio Gardane. / [rule] / m. d. xxxxvi. / CANTVS

The other partbooks have the same title, with changed part-name.

This volume has the same collation as that of the first edition (see note 33, above). Not paginated; the pieces are numbered, i–xxix.

Copy consulted: Munich, Bayerische Staatsbibliothek.

⁴¹ RISM 1543²⁰ = A1377; New Vogel 159.

[flower] CANTVS [flower] / IL TERZO LIBRO D'I MADRIGALI D'ARCHADELT / A QVATRO VOCI INSIEME ALCUNI DI CONST. FESTA / & altri dieci a voci mudate novamente ristampato & corretto. / A QVATRO [Gardane's mark] VOCI / Venetijs Apud Antonium Gardane. / [rule] / m. d. xxxxiii.

The lower voices carry the same title, with changed part-name.

Oblong quarto, 40 pages per part: A–E⁴; F–K⁴; L–P⁴; Q–V⁴. Paginated from the first recto, [1], ii–xxxix, [40], in each partbook.

Copy consulted: Glasgow, Euing Music Library, lacking the last folio of the Bassus.

⁴² These editions, 1539²³ and 1541¹¹, are detailed in note 12, above.

⁴³ This recurrence of identical captions on both formes of a single sheet tends to argue, as I discuss elsewhere in this paper, that some volumes were set more slowly, using only one or two formes for the work. This may be yet another reflection of Gardane's apparent habit of working on more than one title at a time.

⁴⁴ For this volume, see note 9, above. The Scotto edition, RISM 1540²⁰, is described in note 26.

⁴⁵ RISM I534¹⁶ = VI220; New Vogel 2868.

The Bassus survives as

Del Libro Secondo / B with the colophon:

Finisce il Secundo Libro de Madrigali di Verdelot, Nuovamente / Stampati, Et per Andrea Anticho intagliati, 7 con / summa diligentia corretti. / [Scotto's mark] / Venetijs Apud Octauianum Scotum / [rule] / M. D. xxxiii.

Oblong octavo, 32 pages: G–H⁸. Pieces numbered, 1–25. Tavola set in two columns, not headed. All but three composer ascriptions are set vertically in the left margin. Copy consulted: Paris, Bibliothèque Nationale, Rés. Vmf. 40(2).

RISM 1536⁷ = VI221; New Vogel 2869.

II secondo Libro de Madrigali di Verdelot / insieme con alcuni altri bellissimi Madrigali di Adriano, 7 di / Constantino Festa, Nuouamente stampati, 7 con / summa diligentia corretti. / m. d. [large:]sxxxvi. / Con Gratia, 7 Privilegio.

The lower voices have variants of a different title:

Del Libro Secondo di Verdelotto / T

Del Libro Secondo di Verdelot. / A

Del Libro Secondo di Verdelot. / B

On fol. H7^v:

[Scotto’s mark] / [short rule] / Venetijs Apud Octauianum Scotum.

Oblong octavo, 32 pages per part: A–B⁸; C–D⁸; E–F⁸; G–H⁸. Pieces numbered, 1–25. Tavola in two columns, headed. Running head: part-name. Ascriptions at the head of the page in some instances.

Copy consulted: Bologna, Civico Museo Bibliografico Musicale, u 309.

RISM 1537¹⁰ = V1222: New Vogel 1870.

II secondo Libro de Madrigali di Verdelotto / insieme con alcuni altri bellissimi Madrigali di Adriano, 7 di / Constantio Festa: Nuouamente stampati, 7 con / somma diligentia corretti. / md [large:]s xxxvii. / Con Gratia, 7 Privilegio.

The lower voices have a consistent title, with relevant part-letter:

Del Libro Secondo de Madrigali di Verdelotto. / [large:]T

On fol. H7^v

[Scotto's mark] / [short rule] / Venetijs Apud Octauianum Scotum.

Oblong octavo, 32 pages per part: A–B⁸; C–D⁸; E–F⁸; G–H⁸. Pieces numbered, 1–25. Tavola in one column, headed and with ascriptions. Running head-line: part-name, extended on the first recto of gatherings B, D, F and H.

Copies consulted: Munich, Bayerische Staatsbibliothek, 8°. Mus.pr. 40/2, lacks the last folio of the Bassus; Paris, Bibliothèque Nationale, Rés. Vmd. 23 (an Altus book); Bologna, Civico Museo Bibliografico Musicale, R 140/1 (a Bassus).

⁴⁶ RISM 1540¹⁸; New Vogel 2884.

LE DOTTE ET ECCELLENTE COMPOSITIONI / DE I MADRIGALI A CINQVE VOCI DA Dl/versi perfettissimi Musici fatte. Nouamente raccolte, / & con ogni diligentia Stampate. / AVTORI. / Di Adriano Vuillaert. & di / Leonardo Barri suo discipulo. / Di Verdelotto. / Di Constantio Festa. / Di Archadelt. / Di Corteggia. / Di Iachet Berchem. / De Ivo, & di Nolet, / Apud Hieronymum Scotum. / [rule] / 1540.

The lower voices have a consistent different title, with the appropriate part-names:

MADRIGALI A CINQVE VOCE DA PIV / Eccellentissimi Musici fatti. Nouamente raccolti, & / con somma diligentia corretti. / TENOR

Oblong quarto, 44 pages per part: A–E⁴F²; a–e⁴f²; Aa–Ee‘Ff²; aa–ee⁴fr²; AA–EE⁴FF². Paginated from the second recto, i–xlii.

Copy consulted: Munich, Bayerische Staatsbibliothek, 4⁰. Mus.pr. 52(7); Wolfenbüttel, Herzog August Bibliothek, 2.11.1–2.11.5 Musica.

For RISM 1541¹⁷, see note 12, above.

⁴⁷ For this volume, see note 12, above.

⁴⁸ RISM 1539²⁴, 1541¹² and 1545¹⁸, described in note 23, above.

⁴⁹ On this work, see Haar, J., ‘Pace non trovo: a Study in Literary and Musical Parody’, Musica Disciplina, 19 (1965), pp. 95–149, especially pp. 110–19.Google Scholar

⁵⁰ This point was used, above, in determining the printer's view of the authorship of Col pensier mai non maculai.

⁵¹ Arcadelt, v, pp. xiii–xx. Hersh, D., ‘Verdelot and the Early Madrigal’ (Ph.D. dissertation, University of California at Berkeley, 1963), p. 289Google Scholar, suggests that Io nol disse giamai was composed by Verdelot. Bridges, ‘Publishing’, pp. 27–8, uses the fact that Tronchi la parca sets a poem of Lorenzo Strozzi to reinforce a suggested link between Arcadelt and the Strozzi family.

⁵² RISM 1530².

Of this edition, only the Altus partbook survives, with (as title):

A Oblong quarto, 36 pages: J–L⁴M⁶. Presumably the Cantus and Tenor books also had four gatherings each.

Copy consulted: Seville, Biblioteca Colombina, 12–1–31(6) (Altus only).

Chapman, ‘Andrea Antico’, no. 64, identifies this with a book cited in Colon's catalogue as printed in Rome in 1530. Cusick, S., ‘Valerio Dorico, Music Printer in Sixteenth-century Rome’ (Ph.D. dissertation, University of North Carolina, 1975), pp. 61 and 81Google Scholar, believes that the volume was printed by Pasoti and Dorico, and suggests there and in subsequent correspondence with the present writer that there was some connection with the Colonna family. See also K. Jeppesen, ‘Die neuentdeckten Bucher der Lauden des Ottaviano dei Petrucci und andere musikalische Seltenheiten der Biblioteca Colombina zu Sevilla’, Zeitschrift für Musikwissenschaft, 12 (1929–30), pp. 73–89.

⁵³ For these editions, see above, note 12.

⁵⁴ Despite Seay's assertion, in Arcadelt, iv, pp. xii–xiii, that these pieces are anonymous in 1543²⁰, both can be assigned to composers. The pieces are Bramo morir and E morta la speranza, and have been discussed above, in section i.

I am not entirely satisfied that Gardane intended to regard the composer's name, Archadelt, as a free running-title in the second edition. As I am about to show, it was certainly retained from gathering to gathering, but this, of course, is not the same thing. However, the question does not affect the present issue.

⁵⁵ For details of this title, see below, notes 60 and 61.

⁵⁶ See above, note 5.

⁵⁷ Details are above, note 45.

⁵⁸ The single surviving partbook that names Verdelot as the composer has been discussed above, in section ii.

⁵⁹ The incorrect use of the word Alius is on fols. A2^V, A3^r, A6^V, A7^r and A8^V. Of the other pages on this forme, the word is replaced on A4^V and A5^r by the word Cantus, and of course would not have been allowed to remain on A1^r, the title-page. The presence of this error, in gathering A of a book, suggests that this forme had been used for part of an Altus book of some previous publication. The additional presence of another error, the retention of the name of de Silva, in a different forme, suggests that both formes came from another volume, and that they represent only part of the total number of formes (probably four) currently in use by the printer – so that some formes might appear first in this volume in a correct state.

The argument that the reappearance of the word Altus might be from the Altus book of the present title is a weak one, for the following reason: it is probable that at least the inner forme used for gathering A (the first of the Cantus) would be preserved as far as possible, for the first gatherings of the other partbooks, for it contained the Tavola. Indeed, the tavola shows the same type and the same setting in all four partbooks, confirming that it was indeed retained for all. However, the setting of the word Altus in gathering A does not correspond to those in the Altus partbook. Further, the retention of the same formes throughout the first gathering of each partbook implies that not all four (or more, if the printer used more) would be needed for the first gathering: this helps to explain why the reference to de Silva need not have appeared until the second gathering of the Cantus, presumably being set up while the first gathering was going through the press.

As a result, the most simple explanation is that the forme preserving a different setting of the word Altus, together with another forme containing a reference to an irrelevant composer, represent elements of two gatherings at the end of work on some previous title, printed immediately before the present one.

⁶⁰ RISM1533² = V1218; New Vogel 2866.

The surviving Bassus is headed with a capital B and has a colophon:

Finiscono li Madrigali de Verdelot. Stampati novamente / in Vinegia per Zovan Antonio 7 i Fratelli da Sa/bio: Ad instantia de li Scotti: 7 per An/drea Anticho da Montona inta/gliati: 7 con somma dili/gentia corretti. / Con Gratia 7 Privilegio. / [Scotto's mark] / m. d. xxxiii.

Oblong octavo, 32 pages, signed G–H⁸. Pieces numbered, 1–28. The tavola is in two columns with no head-line. A running head-line gives the part-name on the recto and the composer's name on the verso. There are errors on G3^r and H7^r_v. Copy consulted: Paris, Bibliothèque Nationale, Res. Vmf. 40(1).

RISM 1537⁹ = V1219; New Vogel 2867.

II primo libro de Madrigali di Verdelotto. / Novamente stampato, 7 con somma dili-/gentia corretto. / md [large:]sxxxvii. / Con Gratia, 7 Privilegio.

The lower voices are differently titled, with their corresponding part-names:

Del primo Libro de Madrigali di Verdelotto. / [large:]T

The last recto of the Bassus has a colophon:

[mark of Scotto] / Venetijs Apud Octauianum Scotum. [with, in the only surviving copy, the manuscript addition of a date:] mdxxxvii

Oblong octavo, 32 pages per book: A–B⁸; C–D⁸; E–F⁸; G–H⁸. Pieces numbered, 1–28. Tavola in one column, and headed. Running head-line on all pages, of the part-name at the outer edge and the composer's name, centred: an extended form is used on the first recto of gatherings B, D, F and H.

Copies consulted: Paris, Bibliothèque Nationale, Rés. Vmd. 22 (Altus); London, British Library, K.8.b.11 (Altus); Oxford, Bodleian Library, Harding (vii) (imperfect Cantus); Bologna, Civico Museo Bibliografico Musicale, u308.

The table of contents offered by New Vogel for this edition seems to have been printed in error, for it bears no resemblance to the copies that I have seen.

⁶¹ [1536]. Cited in RISM at 1537⁹ = V1219, and in New Vogel at 2867.

The Bassus partbook is entitled:

Del primo Libro de Madrigali di Verdelotto. / [large:]B

Oblong octavo, 32 pages: G–H⁸. Pieces numbered, 1–28. Tavola in two columns, no headline. Running head-line on all folios from G2^r, as in the edition 1537⁹.

Copy at Florence, Biblioteca Nazionale Centrale, Mus.ant. 229. Bassus only, misbound, and lacking H8.

⁶² Unfortunately, the surviving Bassus book lacks the last folio, which would (if the normal procedure of Scotto and Antico was followed here) have carried at least Scotto's mark, and probably a colophon line, giving the date of printing. However, a few details of the analysis should suffice to demonstrate the dating of this volume – independently of its proposed close connection with 1536⁷.

The state of the music blocks is already showing deterioration, though by no means as badly as in the ‘second’ (now known to be the third) edition of 1537⁹. An excellent example is at the end of the last stave in the book, on fol. H7^V. But other obvious cases can be found on G2^V, G6^r, H1^r or H4^V, to mention only one page in each forme of the book.

Several of the initial letters, showing progressive deterioration from 1533 to 1538, can be used to date this edition. Some are particularly vulnerable – and particularly valuable for analysis: the J that appears here on fol. G4^V can be clearly seen to be in a worse condition than that displayed in 1534¹⁶, 1535⁸ or 1535⁹, and in much the same state as that found in 1536⁷; the M on fol. G3^r is also worse than in 1534¹⁶ or 1535⁸, though it is close to the example on H3^V of 1535⁸; the Gon H1^v is again worse than in 1534¹⁶, but it is marginally better than in 1536⁷ and markedly better than in RISM V1224 (also dated 1536), while unfortunately not appearing in the other 1535 editions; of the different forms of the letter L, that found on H6^r is very similar in state to its appearance in 1536⁷, while that on H6^V corresponds in condition to its state in V1223 (the first volume of Verdelot's madrigals à5); the letter S in the new edition is in fact a new letter, which is next to be found only in 1536⁷ and later, while the Von H1^r is in a state close to that of 1534¹⁶ and 1535⁸, markedly better than in 1536⁷ and the succeeding V1224.

The sequence of printing that emerges from this is clear, and incidentally provides an admirable confirmation of the RISM date of [1535] for V1223: in 1535, RISM 1535⁸ precedes 1535⁹ and V1223, all to be followed by the new edition; then 1536⁷ precedes V1224, and 1537¹⁰ probably precedes 1537⁹and 1537¹¹. It is therefore likely that the newly recovered edition of the first book of Verdelot madrigals should be dated [1536], pending the discovery of more parts, or of a Bassus part carrying a colophon.

⁶³ See Bertolotti, A., Musici alla corte dei Gonzaga in Mantova… (Milan, [1890], reprinted Geneva, 1978), p. 34.Google Scholar

⁶⁴ RISM 1542¹⁷.

[flower] D. AVTORI [flower] / IL PRIMO LIBRO D'l MADRIGALI DE DIVERSI ECCELLENTISSIMI AVTORI A MISVRA DI BREVE / NOVAMENTE CON GRANDE ARTIFFICIO / COMPOSTI ET CON OGNI DILIGENTIA / STAMPATI ET POSTI IN LVCE. / QVATVOR [Gardane's mark] VOCVM. / CON GRATIA ET PRIVILEGIO. / Venetijs Apud Antonium Gardane. / [rule] / M. D. xxxxii.

The lower voices have a different title, each with its relevant part-name:

TENOR / QVATVOR [Gardane's mark] VOCVM. / CON GRATIA ET PRIVILEGIO. / Venetijs Apud Antonium Gardane. / [rule] / M. d.xxxxii. [with the addition of a signature line:] Madrigali primi de diuersi autori a 4 [plus signature letter] Oblong quarto, 40 pages per part: A–E⁴; F–K⁴; L–P⁴; Q–V⁴. Paginated in roman numerals from the first recto of each voice-part: [hi], iii–xxxix, [xl].

Copy consulted: Verona, Società Accademia Filarmonica, Busta 208; gathering R is misbound.

⁶⁵ Gardane's shop certainly had more skeleton formes available, and, in any case, would not have wanted to have a press frequently standing idle. It may be that each typesetter could be working with fewer formes than normal whenever the shop was printing more than one title concurrently.

⁶⁶ The sequence of transfer from one forme to another would be as follows: inner forme of gathering A (2^r, 1^v, 4^r, 3^V) to the corresponding outer forme (1^r. 2^V, 3^r;, 4^V), and thence to the inner forme of gathering B; as a result, improbable-looking sequences can occur, such as B4^r to B3^r, or B1^r;to C2^r;.

⁶⁷ Lewis, , ‘Antonio Gardane’, pp. 317–26Google Scholar, and Bernstein, , ‘Burning Salamander’, p. 500.Google Scholar

⁶⁸ The evidence that might seem to be offered by the Florentine manuscript, Florence, Biblioteca Nazionale Centrale, Magi, xix 122–5, is of little value if, indeed, the source is largely derived from printed editions.

⁶⁹ Nugent, G. and Haar, J., ‘Maistre Jhan’, The New Grove Dictionary of Music and Musicians, ed. Sadie, S., 20 vols. (London, 1980), xi, p. 541.Google Scholar

⁷⁰ An exception would be RISM [c1530]¹_; the Libro primo de la fortuna, where we cannot tell whether the attributions stop mid-way down the tavola because the printer left his anonyma to the end, or because he lost interest.

⁷¹ Bridges, ‘Publishing’, p. 43, adduces some evidence for suggesting that Arcadelt may have been in Rome in 1538. Certainly, he argues well that Arcadelt was probably not still in Florence.

⁷² Slim, H.C., ‘Verdelot, Philippe’, The New Grove Dictionary, xix, p. 633.Google Scholar See also J. Haar, writing of Verdelot's influence on the development of the madrigal, and of the two extremes of style that had existed: ‘he may be thought to have followed, and occasionally combined, both trends’. This appears in his ‘The Early Madrigal: a Re-Appraisal of its Sources and its Character’, Music in Medieval and Early Modern Europe: Patronage, Sources and Texts, ed. Fenlon, I. (Cambridge, 1981), p. 78, fn.Google Scholar

Addendum to note 25: Bernstein, J., ‘The Burning Salamander: Assigning a Printer to some Sixteenth-century Music Prints’, Notes, 42 (1985–1986), pp.483–501CrossRefGoogle Scholar (which appeared after this paper was completed), clarifies many points regarding these volumes. She leaves open the question why the interconnected group of printers and patrons should wish (most unusually among partnerships) to leave the volumes unassigned, as also any reason why these particular books should be involved. My discussion needs to be read in conjunction with her analysis, and with further detailed study of editorial practice.