By either of these methods negatives of extreme density and absolute clearness in the lines may be produced and used for obtaining impressions in ink, the transfer of which to stone presents no difficulty. It remains to be added that there are several methods of producing grained negatives for photo-lithography in half-tone, and therefrom a transfer for stone, and so obtaining impressions at a much greater rate than is possible from a Collotype plate. They may each produce results with a certain amount of success; but even those from the best methods indicated cannot be compared with Collotype. The too obtrusive grain of the half-tone pictures so produced destroys much of the finer details, and it is to be regretted that all such processes leave so much to be desired. APPENDIX. THE STEAM COLLOTYPE MACHINE. The earliest steam machines for Collotype printing were made by Faber & Co., of Offenbach; they were, however, so complicated and delicate in construction that other machines have entirely supplanted them. The three principal varieties are those of Alauzet & Co., of Paris; Koch, of Leipzig; and Schmiers, Werner & Stein, each of which may be considered entirely successful; they possess features of great similarity, differing not in principle, but only in details of construction. Contrasting the work of the machine with that of the press, the printing is much more rapid, and the prints more uniform and regular, while the “life” of the plate is greatly lengthened; one capable of yielding 250 impressions on the scraper press would, in all probability, give 1,000 on the machine—a fact easily accounted for when it is considered that the one is produced by a scraping pressure while the other is made by the rotation of a cylinder. As will be seen by the illustrations, the machines bear a considerable resemblance to those used in lithography—they may, in fact, be used for lithographic printing—with very greatly increased inking power. The machines are so constructed that the plate may be inked once, twice, or three times, each inking giving more vigour, but, of course, a corresponding decrease in the number of impressions yielded; the output may be relatively estimated at 1,500, 1,000, and 800 per day. It is, of course, an impossibility to teach machine printing by means of instructions such as may be conveyed in a work like the present, and the following remarks are added more with a view of giving a general idea of the nature of the necessary operations than with any idea of making the reader a competent machine minder.