If the “nice things” were matrices, there must have been a type-mould, and it was this mould which was the key to the invention. The mould was the only implement connected with typography which would at once lay open to an intelligent observer the secret of making types. Of all his tools, this was the one that had received the greatest amount of care and labor, and it should have been the one that Gutenberg would be anxious to conceal. It may be supposed that the thing of four pieces that was opened by two buttons was the mould.[245] Why it should have been kept in or under the press cannot be explained. But if Dritzehen was fitting up matrices, it was proper that he should have the mould at hand. The conjecture that the thing of four pieces was a type-mould, is not free from difficulties, but it seems the only one that makes intelligible the action of the witnesses. [p399] [Illustration: Fac-simile of the Type-mould of Claude Garamond. _a._ The place where the body of the type was cast. _b. c._ The mouth-piece in which the fluid metal was poured. _d._ The type as cast, with the metal formed in the mouth-piece adhering to it. [From Duverger.]] The gravest difficulty in the way of this conjecture is, that the type-mould of modern type-founders has, including the matrix, but three detachable pieces. As this mould is substantially of the same form as that known to have been used by Claude Garamond, the eminent type-founder of Paris, in 1540, it has been supposed, and properly, that this mould of three pieces must have been used before Garamond, by all the early printers. But it was not the only form of mould. At the beginning of this century every type-founder found it expedient to use at times, a type-mould somewhat different in its construction—a mould which, with the matrix, consisted of four detachable pieces. The merit of this mould was its adaptability, within limits, to any size of body. Its disadvantages were its difficulty of nice adjustment and its liability to inaccuracy—faults which have obliged all American type-founders of this day to discontinue its use entirely. It is, without doubt, a very old form of mould, but it was never a popular one, having been used chiefly for casting bodies of irregular size.[246] Mr. Bruce has showed me one of these early moulds—a mould long out of use, preserved only as one of the earlier relics of his old type-foundry. Its construction is [p400] too complex for description by words, or even by engraving; but it may be sufficient to say that, with the matrix, it consisted of four pieces, and was so constructed as to allow of an enlargement and nice adjustment in either direction of the space provided for casting the body of the type. The pieces were held together by stiff springs, but buttons could have been used for the same purpose. When these pieces were connected it would be plain to any mechanic that it was a mould; disconnected, its purpose would be a riddle.