exhibited by certain Non-metallic Substances. 401 



the very remarkable mode of absorption of light by red solution 

 of mineral chameleon*^. This solution, which may be regarded 

 as an optically pure solution of permanganate of potash, inas- 

 much as it is associated with only a colourless salt of potash, 

 absorbs green light with great energy, as is indicated by the 

 tint, even without the use of a prism. But when the light 

 transmitted by a pale solution is analysed by a prism, it is found 

 that there are five remarkable dark bands of absorption, or 

 minima of transparency, which are nearly equidistant, and are 

 situated mainly in the green region. The first is situated on 

 the positive or more refrangible side of the fixed line D, at a 

 distance, according to a measurement recently taken, of about 

 four-sevenths of the interval between consecutive bands ; the last 

 nearly coincides with F. The first minimum is less conspicuous 

 than the second and third, which are the strongest of the set. 

 Now it occurred to me, that as the solution is so opake for rays 

 having the refrangibilities of these minima of trans parency, cor- 

 responding maxima might be expected in the light reflected from 

 the crystals. This expectation has since been realized by obser- 

 vations made on some small crystals. On analysing the reflected 

 light by a prism, I was readily able to observe four bright bands, 

 or maxima, in the spectrum. These, as might have been expected, 

 were more easily seen when the light was incident nearly per- 

 pendicularly than at a large angle of incidence. The first 

 band was yellow, the others green, passing on to bluish-green. 

 On decomposing the light reflected at a considerable angle of 

 incidence, in a plane parallel to the axis, into two streams, 

 polarized respectively in, and perpendicularly to, the plane of 

 incidence, and analysing them by a prism, the bands were hardly 

 or not at aft perceptible in the spectrum of the former pencil, 

 while that of the latter consisted of nothing but the bright bands. 



The tint alone of the first bright band already indicated that 

 the band was more refrangible than the light lying on the nega- 

 tive side of the first dark band seen in the spectrum of the light 

 transmitted by the solution, and less refrangible than the light 

 lying between the first and second dark bands, so that its position 

 would correspond, or nearly so, to the first dark band. However, 

 the eye is greatly liable to be deceived, in experiments on absorp- 

 tion, by the effects of contrast, and therefore an observation of the 

 nature of that just mentioned cannot, I conceive, be altogether 

 relied on. The smallness of the crystals occasioned some diffi- 

 culty ; but a more trustworthy observation was made in the fol- 

 lowing manner. 



The sun's light was reflected horizontally into a darkened 



* Phil. Trans, for 1852, p. 558. 



