[22] It is well known that until comparatively recently the large “proscription letters” of our foundries, from three-line pica and upwards, were cast in sand. The practice died out at the close of last century.
[23] An Enquiry Concerning the Invention of Printing. London, 1863, 4to, p. 265.
[24] In a recent paper, read by the late Mr. Bradshaw of Cambridge, before the Library Association, he points out a curious shrinkage both as to face and body in the re-casting of the types of the Mentz Psalter, necessary to complete the printing of that work. The shrinking properties of clay and plaster are well known, and, assuming the new type to have been cast in moulds of one of these substances formed upon a set of the original types, the uniform contraction of body and face might be accounted for. If, on the other hand, we hold that the types of this grand work were the product of the finished school of typographers, the probability is that the new matrices (of the face of the letter only) were formed in clay, as suggested at p. [15], and that the adjustable mould was either purposely or inadvertently shifted in body to accommodate the new casting.
[25] In connection with the suggested primitive modes of casting, the patent of James Thomson in 1831 (see Chap. iv, post), for casting by a very similar method, is interesting.
[26] Origine de l’Imprimerie. Paris, 1810, 2 vols., 8vo, i, 97.
[27] Origine de l’Imprimerie, i, 99, etc. The following are the citations:—“Escriture en molle,” used in the letters of naturalisation to the first Paris printers, 1474. “Escrits en moule,” applied to two Horæ in vellum, bought by the Duke of Orleans, 1496. “Mettre en molle,” applied to the printing of Savonarola’s sermons, 1498. “Tant en parchemin que en papier, à la main et en molle,” applied to the books in a library, 1498. “Mettre en molle,” applied to the printing of a book by Marchand, 1499. “En molle et à la main,” applied to printed books and manuscripts in the Duke of Bourbon’s library, 1523. “Pièces officielles moulées par ordre de l’Assemblée.” Procès verbaux des Etats Généraux, 1593.
[28] Coster Legend, p. 6.
[29] Ibid., p. viii.
[30] A calculation given in the Magazin Encyclopédique of 1806, i, 299, shows that from such matrices 120 to 150 letters can be cast before they are rendered useless, and from 50 to 60 letters before any marked deterioration is apparent in the fine strokes of the types.
[31] Several writers account for the alleged perforated wooden and metal types reputed to have been used by the first printers, and described by Specklin, Pater, Roccha and others, by supposing that they were model types used for forming matrices, and threaded together for safety and convenience of storage.