¹ The felspar-fragments in these deposits are not well suited for identification of varieties, as they are to a great extent quite irregular, ragged bits, few of them showing any definite boundaries or cleavages with which to correlate extinctions and optic figures. The larger bits () are mainly more or less turbid, like most of the felspars in the rocks from which they are derived; but among the smaller grains of the more finely-sifted materials there are plenty of clear fresh bits to be found.

It may be remarked,inthis connection, that even with very imperfect material, both in a fragmental condition and in rock-sections, a good deal more can be done in the way of identification than is often supposed, by making use of Schuster's method of extinctions,inconjunction with a careful application of the figures in convergent polarized light. It is often not necessary to have cleavage-flakes sufficiently good to enable the sign as well as the angle of extinction to be determined. Thus, suppose we have a very imperfect flake, or an irregular bit in a slide, showing, however, one good cleavage-crack or boundary. If we find the extinction is 18°–20°, measured against this line, and if we have at the same time an undoubted emergence of a positive bisectrix not quite central in the field, then we know that we have albite. a similar fragment or bit in a section, giving an extinction of about 36°, with a sharp and distinct axial bar well within the field, but near its edge, will be anorthite. Other determinations may be made in this way on very unpromising material, provided the optic figures are understood and are carefully used as a guide, and no attempt is made to draw inferences except when these figures can be determined beyond question.

This is not an appropriate place in which to go more fully into this subject, which can be easily followed out by studying the details given in Rosenbusch, orinTeall's “British Petrography” or other works, as to Schuster's method, and the optic figures on the M flakes of the different felspars.

To obtain good results it is necessary, as I pointed out in previous remarks on the determination of felspar (GEOL. MAG. January, 1894, p. 41), to have efficient optical apparatus.invery thin flakes, or in bitsinthin sections, owing to the very low polarization-tints, it is often difficult to determine the extinctions with sufficient accuracy. For this purpose a Bertrand quartz-eyepiece (with four quartz segments) will be found of the greatest service.

The discrimination of the felspars is, of course, of the highest importance in all petrological work, and especially so in the study of metamorphic rocks, so that too much stress cannot be laid upon the desirability of carrying it out in all cases as fully and as definitely as the nature of the material will allow. It appears to be sometimes left undone, or very vague, where there is no necessity to so leave it; or it is even left undecided whether a mineral is quartz or felspar,—an uncertainty which is seldom necessary where the mineral or minerals occur in any reasonable quantity in a slide.