One thinks of El burlador de Sevilla and Tan largo me lo fiais; these are often said to be independent descendants of a lost original, but if they are, they vary so greatly that it will be almost impossible to reconstruct the original.

Lope's El castigo sin venganza survives in an autograph manuscript of 1631 and in an edition, apparently prepared by Lope, of Barcelona, 1634; they vary. Calderón's El postrer duelo de España survives in three manuscripts, one of which has portions in Calderón's hand, as well as in editions of 1672 and 1674, in the second of which the plays are allegedly «enmendadas, y corregidas». Again, the versions vary.

See H. A. Rennert, The Spanish Stage in the Time of Lope de Vega (New York, 1909), pp. 177-8, for prices paid by stage managers. It is hard to discover what a publisher would have paid for a new play, since new plays were almost never sold for publication. William Byron, in his Cervantes: A Biography (London, 1979), pp. 414, 482, quotes some of the fees paid by publishers for prose works (Cervantes's Novelas ejemplares and Galatea, Rojas Villandrando's Viaje entretenido and Espinel's Marcos de Obregón): they range from 1100 to 1600 reales. If comparable sums were paid for volumes of plays, their value to a publisher was about 90 to 130 reales each. The price paid for Cervantes's Ocho comedias, y ocho entremeses nuevos, nunca representados is not known, but is reckoned to have been much less than the 1600 reales he got for the Novelas ejemplares.

One unambiguous example of a late revision is El mayor monstruo los celos, printed in 1637, but surviving in a manuscript in which Calderón's revision can be dated by handwriting changes to circa 1670.

For further information on how books were made, see P. Gaskell, A New Introduction to Bibliography (Oxford, 1972); for the terminology of descriptive bibliography, see F. Bowers, Principles of Bibliographical Description (Princeton, 1949).

Gaskell, A New Introduction, p. 42 (Plantin); D. W. Cruickshank, «Some Aspects of Spanish Book-production in the Golden Age», The Library, V, 31 (1976), 3, n. 11. To distinguish between setting by formes and setting by pages, one must check the recurrence of identifiable pieces of type; see D. W. Cruickshank, «The Printing of Calderón's Tercera parte», SB, 23 (1970), 230-51.

The easiest way to identify the format of a Spanish book is by its size. Almost all ordinary Spanish books were printed on foolscap, which was about 44 by 31 cm. Each successive fold halved the longer dimension, so that folio was 31 by 22, quarto 22 by 15.5, octavo 15.5 by 11cm. After trimming, these dimensions would be 30 by 21, 21 by 14.5 and 15 by 10.5 cm. approximately. Thus a folio will be about 12 inches tall, a quarto just over 8 inches and an octavo not quite 6 inches. If doubt still remains, the paper must be examined to establish the direction of the chain-lines and the position of the watermarks (if any): see Gaskell, A New Introduction, pp. 60-77, 84-6. The final gathering of a quarto in eights will often consist of more or less than two complete sheets, e. g. Z6, Z10, Z12; that is, 1½ sheets, 2 ½ sheets, 3 sheets. Investigators will find it easy to draw diagrams of the formes involved if they remember that the half sheet was always innermost in the gathering. Thus the outermost sheet of a Z10 gathering will have 1v, 2r, 9v, 10r as its inner forme, 1r, 2v, 9r, 10v as its outer. The middle sheet will have 3v, 4r, 7v, 8r and 3r, 4v, 7r, 8v. The innermost (half) sheet will have 5v, 6r and 5r, 6v.

For more details on proofing, see Gaskell, A New Introduction, pp. 110-16.

For press-variants in Spanish plays, see Cruickshank, «The Printing of Calderón's Tercera parte», 238; and «The Text of La vida es sueño», in The comedias of Calderón, edited by D. W. Cruickshank and J. E. Varey, 19 vols. (Farnborough, 1973), I, The Textual Criticism of Calderón's comedias, pp. 79-87. The variants described in these articles were discovered by means of a Hinman collating machine, which can collate only originals and good original-size photocopies. More modern aids to collating use TV cameras and can cope with microfilms: John Horden, «The Institute of Bibliography and Textual Criticism, Leeds», The Library, V, 27 (1972), 300-1.

On the documents of the preliminaries, see J. Moll, «Problemas bibliográficos del libro del Siglo de Oro», BRAE, 59 (1979), 51-7, which also deals briefly with the practice in other Spanish kingdoms; and J. Simón Díaz, La bibliografía: conceptos y aplicaciones (Barcelona, 1971), pp. 127-72. Moll's is perhaps the best general introduction to the circumstances peculiar to the production of Spanish books of this period.