and wiry letter of Plantin and the Elzevirs having served for the model, rather than the generous and logical designs of the fifteenth-century Venetian printers, at the head of whom stands Nicholas Jenson; when it is so obvious that this is the best and clearest roman type yet struck, it seems a pity that we should make our starting-point for a possible new departure at any period worse than the best. If any of you doubt the superiority of this type over that of the seventeenth century, the study of a specimen enlarged about five times will convince him, I should think. I must admit, however, that a commercial consideration comes in here, to wit, that the Jenson letters take up more room than the imitations of the seventeenth century; and that touches on another commercial difficulty, to wit, that you cannot have a book either handsome or clear to read which is printed in small characters. For my part, except where books smaller than an ordinary octavo are wanted, I would fight against anything smaller than pica; but at any rate small pica seems to me the smallest type that should be used in the body of any book. I might suggest to printers that if they want to get more in they can reduce the size of the leads, or leave them out altogether. Of course this is more desirable in some types than in others; Caslon's letter, e. g., which has long ascenders and descenders, never needs leading, except for special purposes. I have hitherto had a fine and generous roman type in my mind, but after all a certain amount of variety is desirable, and when you have gotten your roman letter as good as the best that has been, I do not think you will find much scope for development of it; I would therefore put in a word for some form of gothic letter for use in our improved printed book. This may startle some of you, but you must remember that except for a very remarkable type used very seldom by Berthelette (I have only seen two books in this type. Bartholomew, the Englishman, and the Gower, of 1532), English black-letter, since the days of Wynkin de Worde, has been always the letter which was introduced from Holland about that time (I except again, of course, the modern imitations of Caxton). Now this, though a handsome and stately letter, is not very easy reading; it is too much compressed, too spiky, and so to say, too prepensely gothic. But there are many types which are of a transitional character and of all degrees of transition, from those which do little more than take in just a little of the crisp floweriness of the gothic, like some of the Mentelin or quasi-Mentelin ones (which, indeed, are models of beautiful simplicity), or say like the letter of the Ulm Ptolemy, of which it is difficult to say whether it is gothic or roman, to the splendid Mainz type, of which, I suppose, the finest specimen is the Schoeffer Bible of 1462, which is almost wholly gothic. This gives us a wide field for