Page:Scientific Memoirs, Vol. 1 (1837).djvu/75

Rh screens of differently coloured glass, in order to make them pass through one common screen of alum.

We see here that the rays emerging from the red, orange, yellow, blue, indigo, and violet are transmitted through the plate of alum in the same proportion as the rays that issue from the white glass. The colouring matter introduced into the composition of these different kinds of glass has no other effect than to extinguish part of the calorific sheaf which passed through the white, without perceptibly altering the relations of quantity between the several species of rays of which that sheaf is composed: they act in respect to radiant heat just as brown or blackish substances dipped in a transparent fluid would act in respect to light. But the case is different with respect to green and opake black; for these being introduced into the composition of glass, it will stop nearly all the rays that the alum is capable of transmitting. This effect arises from the green or opake colouring matter producing a certain modification in the diathermancy of the glass, and we have just seen that this species of calorific colouration is invisible and totally independent of coloration properly so called, since it exists in bodies possessing the greatest transparency. It is then extremely probable that the black or the green should not be supposed to enter as mere neutrals into this phænomenon, which will thenceforth depend on which or such a property of these colouring materials. I have found, in fact, some green glasses, which produced a much feebler action than others of the same tint but possessing a less brilliant colouration. The green glasses which act most powerfully are of a bluish cast; from which circumstance it would seem to follow that they contain a considerable quantity of oxide of copper. Whatever may be said of this singular property of green and black opake glasses and the cause by which it is produced, it is nevertheless an indisputable fact which every man can easily verify and of which we intend to give some new proofs presently. But it will perhaps be advisable previously to adduce the results furnished by several