A 3/4 B 1-1/2 C 2-3/8 D 3-1/2 E 5-11/16 b 6-3/16 F 7-1/2 G 10-5/8 At first plates of selenite, which are easily prepared, were used, because they can be split to nearly the requisite thickness with parallel faces; but their depolarising power varied much with temperature. Even the ordinary atmospheric changes alter the position of the bands. However, quartz cut parallel to the principal axis of the crystal is but slightly affected, and is not open to the same objection; but this is prepared with some difficulty. The sides should be perfectly parallel, the thickness about ·043-inch, and gradually polished down with rouge until the sodium-line is seen in its proper place. This must be done with care, since a difference of 1/10000-inch in thickness would make it almost worthless. The two Nicol’s prisms and the intervening plate are mounted in a tube, and attached to a piece of brass in such a manner that the centre of the aperture exactly corresponds to the centre of any of the cells used in the experiments, and must be made to correspond with equal care, so that any of them, or this apparatus in particular, may be placed on the stage and in proper position without further adjustment, whereby both time and trouble are saved. Absorption Spectrum of Chromule. In 1869 I published in the Journal of the Royal Microscopical Society[38] a paper on results obtained by the spectrum analysis of the colouring-matter of plants and flowers, some of which were of considerable interest in many respects. My examinations extended to several hundred different specimens, from which I was led to conclude that the chromule of flowers is, for the most part, due to the chemical action of the actinic rays of light over the protoplasm of the plant, more so than to that of soil. But as certain roots of plants, as those of the alkanet, yield their colouring-matter to oil, and in a much smaller degree to spirit or water, it follows then that conclusions of any kind can only be drawn after a long and careful study of the question. Some of the results obtained were, however, of some interest at the time, that, for example, seen in three different solutions of the chlorophyll of _Cinchona succirubra_, one of three solutions in alcohol, scarcely coloured, having in fact only a faint tinge of green colour, and the spectrum of which much astonished me at the time. It gave four well-marked absorption-bands; one deep sharp line _in the red_; another, rather narrower, in the orange, coincident with D, or the sodium-line; one in the green, about _b_, coincident with the Thallium green band; and a fourth on the blue line F, nearly as broad as that in the red. The ethereal solution gave different results. It showed only three bands of absorption, nearly the same as in the last case (though all of them fainter); but the fourth in the blue was not apparent, the whole of that end of the spectrum being absorbed a little