When, therefore, the printer of the _Catholicon_, in 1460, says of his book, “non calami styli aut pennæ suffragio, sed mirâ patronarum formarumque concordiâ proportione ac modulo impressus atque confectus est,” we have not necessarily to conclude that the types were produced in the modern way from copper matrices struck by steel punches. Indeed, probability seems to point to a gradual progress in the durability of the materials employed. In the first instance, the punches may have been of wood, and the matrices soft lead or clay[36]; then the attempt might be made to strike hard lead into soft; that failing, copper punches[37] might be used to form leaden matrices; then, when the necessity for a more durable substance than lead for the letter became urgent, copper would be used for the matrix, and brass, and finally steel, for the punch. Of whatever substance the matrices were made, the first printers appear early to have mastered the art of justifying them, so that when cast in the mould they should not only stand, each letter true in itself, but all true to one another. Nothing amazes one more in examining these earliest printed works than the wonderful regularity of the type in body, height, and line; and if anything could be considered as evidence that those types were produced from matrices in {17} moulds, and not by the rude method of casting from matrices which comprehended body and face in the same moulding, this feature alone is conclusive. We may go further, and assert that not only must the matrices have been harmoniously justified, but the mould employed, whatever its form, must have had its adjustable parts finished with a near approach to mathematical accuracy, which left little to be accomplished in the way of further improvement.