181
See p. 36
182
Both, moreover, do this when there are several other works in the Opera which could have been used and were not, and when neither De Rebus familiaribus nor Bucolicum Carmen was an obvious choice as a source for a Spanish writer at the end of the fifteenth century. A further resemblance is the preference of both parts of La Celestina for Book I of De Remediis.
183
G. Delpy, 'Les Profanations du texte de La Celestina', Bulletin Hispanique, xlix (1947), 261-75
184
Gilman, op. cit., 169-70, maintains that the absence of Petrarch from Act I (apart from the borrowing which he claims to have established) is due to the 'lack of thematic need rather than to mere lack of knowledge'; he feels that the nature of Act I and its place in the structure of La Celestina are such that Petrarchan borrowings would be out of place, and he dismisses Castro Guisasola's perfectly correct statement that there are no Petrarchan borrowings in the act. This explanation would be sufficient if Petrarchan borrowings elsewhere in La Celestina were all of one type, or were all used in relation to one theme, but their diversity is such that it is difficult to imagine a set of circumstances which would exclude them all from 86 pages.
185
F. Vindel, El arte tipográfico en España durante el siglo XV, vols. vi (Madrid, 1950), 130-3, and vii (1951) xxv-xxvi, regards the Burgos? 1499? edition as a sixteenth-century production, and believes that the edition of Toledo, 1500, regarded by most bibliographers as the second extant, is princeps.
186
See p. 67. This slip is the more remarkable in that Rojas had to overlook the presence of Nicomedes in the Petrarchan passage. The rendering of Prusias as Bursia is also interesting.
187
See p. 42. This is noted by Cejador.
188
Op. cit. 76-78. This is not the only case of an apparent printer's error: R. Menéndez Pidal (Antología de prosistas españoles, 5ª ed., Madrid 1928-RFE, Anejo ii-p. 69) suggests the correction of 'Eras y Crato médicos... piedad de silencio' (i. 35-6; 25), which is found in the 1499? edition, to 'Erasistrato... piedad de Sileuco'; other emendations have been proposed. Otis H. Green, 'Celestina, Auto I. Minerua con el can', NRFH, vii (1953),470-4, corrects el can to Vulcán.
189
The proportion of the errors (3 in 1499?, 1 in 1502) corresponds nearly enough to the proportion of borrowings in the two versions. There is occasional clumsiness in translation, but this again is infrequent. An example is the rendering 'circulatorum ludus' as 'juego de hombres que andan en corro' (see p. 73).
190
The intensification in 1502 of stylistic features found in the original version has been noted by several critics.