Calderón's Tercera parte of 1664 was set by formes (D. W. Cruickshank, «The printing of Calderón's Tercera parte», SB, 23, 1970, 230-51); so, apparently, was his Quarta parte of 1672. In 1675 Cabrera said of compositors that «they have all to be very expert at... casting off any original, because books are not set seriatim, but by alternating [passages of] the original, in order to produce two, three or four sheets placed inside each other» (Discurso, fol. 15r). Plantin, who had plenty of type, was setting by pages in the sixteenth century (Gaskell, A new introduction, p. 42). Setting by formes was apparently common enough in England in the late sixteenth and early seventeenth centuries (e. g. G. W. Williams, «Setting by formes in quarto printing», SB, 2, 1958, 39-53), but Moxon did not recommend it in 1683 (J. Moxon, Mechanick exercises, ed. H. Davis and H. Carter, 1962, pp. 210-11).

E. M. Wilson, «The two editions of Calderón's Primera parte of 1640», The Library, V, 14» 1959. 175-91.

The vast quantity of this ephemeral material, and the failure of modern literary historians to take account of it, is brilliantly described by Don Antonio Rodríguez-Moñino in Construcción crítica y realidad histórica en la poesía española de los siglos XVI y XVII, Madrid, 1965. See also E. M. Wilson, Some aspects of Spanish literary history: the Taylorian Lecture, delivered 18 May 1966, Oxford, 1967, pp. 19-20.

One factor which possibly led to a greater amount of piratical printing in Spain was the judicial separation of Aragón and Portugal from Castile, which meant that a privilege granted for Castile was not valid in Aragón or Portugal (and vice-versa). Aragonese printers had a particularly bad reputation for pirating books covered only by a Castilian privilege. Castilian printers retaliated by producing books with false Aragonese imprints, e. g. Calderón's Quinta parte of «Barcelona», Madrid, 1677.

The Dawson catalogue lists 1,717 items printed in Spain (modern boundaries) with no imprint. Of the 1,717, 1,236 (72 per cent) are assigned to Madrid. The other 481 are divided among thirty-five towns and cities, a most improbable distribution when one considers that Madrid accounts for only about 40 per cent of the imprinted items.

The publication of Norton's complete catalogue of Spanish books from 1501-20 will complement his Printing in Spain 1501-1520, Cambridge, 1966, and give us an excellent knowledge of Spanish printing until 1520. From then on, until well into the eighteenth century, our knowledge is more and more fragmentary with the advance in date. The most formidable collection of documents on early Spanish printing is that of J. M. Madurell y Marimón and J. Rubió y Balaguer, Documentos para la historia de la imprenta y librería en Barcelona (1474-1553), Barcelona, 1955.

Correspondance de Christophe Plantin, ed. M. Rooses and J. Denucé, Antwerp, 1883-1918: iv, 1914, 117-19 (letter 546), 3 August 1574.

His hebrew may be seen in the Commentaria in Isaiam of León de Castro (1570), used in the text on an apparent body of 96 mm.; but it appears to be used on †4v on an 81 mm. body. The greek is also present in the text (96 mm.), but on 2†4v-6r on a 67 mm. body. The pica italic («Proben's second») appears as pica on §3r (81 mm.) of the same work, but as «english» (96 mm.) on p. 95. Oddities in the alignment of the print in the Melchor Cano (p. 73) and in a Dioscorides of 1566 suggest that the use of the long primer greek as english was achieved by leading; but the english roman Gast had in 1563 appears in 1570 on the same body (96 mm.) with a wider «set», which doubtless indicates a new fount cast from the same matrices in a different (or slightly adjusted) mould.

«buen fundidor digo buen fundidor por que no basta mal fundidor que destos ay aca hartos syno uno bueno que sepa ajustar las matrices y hazer moldes» (letter 546).

In his reply (Correspondance, iv, 148-9: letter 562, 4 October 1574) Plantin says that he printed two editions of the work Gast refers to. These must be Alan Cope's Dialogi sex contra Summi Pontificatus... oppugnatores, printed in 1566 and 1573. If Gast intended the first, he wanted Garamont's gros romain romain and english roman; if the second (which is perhaps more likely, since he was writing the year after it appeared), Garamont's english roman and pica roman. In either case he evidently wanted Garamont's english roman, so Plantin's remark about two editions was a red herring. I have seen none of the three faces in Gast's books.