In [50] and [51] may be found an attempt to formulate a scheme to assist in the reconstruction of an alphabet of Roman small letters, after somewhat the same fashion as

that devised for the Roman capitals by Mr. Ross, in [1] and [2]. A small-letter diagram must, for obvious reasons, be less exact and detailed than one for the more defined capital form; but the diagram given will serve to determine sufficiently the main outlines and proportions. In their shapes the letters shown in [50] and [51] adhere fairly closely to the best type forms of the small letter; and the drawing will serve, further, to show the space generally allowed by modern founders between one lower-case letter and another when set into type words. This spacing is based on the m of the fount employed. The open space between all but k, w and y (in which the outlines of the letters themselves hold them further away from their neighbors) and the round letters being the space between the upright strokes of the m; an interval represented in the diagram by a square and a half. The round letters, as has already been said in speaking of the capital forms, should be spaced nearer together; and it will be observed that they are only separated by one square in the diagram. Although suggestive, the rules which govern the spacing of types are not to be blindly followed by the pen letterer. In type, for instance, it would be impossible, for mechanical reasons, to allow the kerns of the f, j and y to project far over the body of the next letter, and in these letters the kerns consequently have either to be restrained or the letters spaced farther apart. In pen lettering, however, the designer is not restrained by such limitations, and his spacing of letters should be governed solely by the effect.

The disposition of the accented lines in the small letters follow the same general rules that govern those of the capitals (see page [2]); the only deviation being in the case of

the g, in which the shading of the bottom seems to have been determined largely by the effect upon the eye.

It will be noticed in the diagram that the "ascenders" of the smaller letters rise about three squares to their extreme top points above the body of the letter; that the body of each letter is inclosed in a square that is three units high, and that the "descenders" fall but two squares below the letter body. These proportions are not by any means invariable, however, and indeed there is no fixed rule by which the proportions of ascenders and descenders to the body of the Roman minuscule may be determined. In some forms of the letter both are of the same length, and sometimes that length is the same as the body height of the letter. In general a better result is obtained by making both ascenders and descenders of less than the length of the body, and keeping the descenders shorter than the ascenders in about the proportion of two-fifths to three-fifths.

Parallel lines of small letters cannot be spaced closer to each other than the ascenders and descenders will allow; the projections above and below the line are awkward, and interrupt the definite lines of demarkation at the top and bottom of the letter-bodies; the capitals necessarily used in connection with the small letters add to the irregularity of the line—all of which reasons combine to limit the employment of minuscule for formal or monumental uses. On the other hand, the small letter form is excellently adapted for the printed page, where the occasional capitals but tend to break the monotony, while the ascenders and descenders strongly characterize and increase the legibility of the letter forms.