This experiment may be made with the prismatic spectrum. The temperature of the room being first remarked on the thermometer, the blue coloured light is made to fall on the bulb, when a somewhat higher degree of warmth is exhibited, which still increases as the other colours are gradually brought to act on the mercury. If the experiment is made with the water-prism, so that the white light can be retained in the centre, this, refracted indeed, but not yet coloured light, is the warmest; the other colours, stand in relation to each other as before. 677. As we here merely describe, without undertaking to deduce or explain this phenomenon, we only remark in passing, that the pure light is by no means abruptly and entirely at an end with the red division in the spectrum, but that a refracted light is still to be observed deviating from its course and, as it were, insinuating itself beyond the prismatic image, so that on closer examination it will hardly be found necessary to take refuge in invisible rays and their refraction. 678. The communication of light by means of coloured mediums exhibits the same difference. The light communicates itself to Bologna phosphorus through blue and violet glasses, but by no means through yellow and yellow-red glasses. It has been even remarked that the phosphori which have been rendered luminous under violet and blue glasses, become sooner extinguished when afterwards placed under yellow and yellow-red glasses than those which have been suffered to remain in a dark room without any further influence. 679. These experiments, like the foregoing, may also be made by means of the prismatic spectrum, when the same results take place. 680. To ascertain the effect of coloured light on oxydation and de-oxydation, the following means may be employed:--Let moist, perfectly white muriate of silver[1] be spread on a strip of paper; place it in the light, so that it may become to a certain degree grey, and then cut it in three portions. Of these, one may be preserved in a book, as a specimen of this state; let another be placed under a yellow-red, and the third under a blue-red glass. The last will become a darker grey, and exhibit a de-oxydation; the other, under the yellow-red glass, will, on the contrary, become a lighter grey, and thus approach nearer to the original state of more perfect oxydation. The change in both may be ascertained by a comparison with the unaltered specimen. 681. An excellent apparatus has been contrived to perform these experiments with the prismatic image. The results are analogous to those already mentioned, and we shall hereafter give the particulars, making use of the labours of an accurate observer, who has been for some time carefully prosecuting these experiments.[2] [1] Now generally called chloride of silver: the term in the original is Hornsilber.--T. [2] The individual alluded to was Seebeck: the result of his experiments was published in the second volume.--T. LVI.