148 SIR J. F. W. HERSCHEL ON THE 



many vessels filled with it. Accordingly it might be supposed that by passing the 

 same incident beam successively through many such dispersive surfaces, the whole 

 of the blue rays would at length be separated from it, and an orange, or red residual 

 beam be left. But this is not the case, the reason of which is to be found in a very 

 remarkable peculiarity in the transmitted light, which may be thus announced. 



An epipolized beam of light (meaning thereby a beam wliich has been once trans- 

 mitted through a quiniferous solution and undergone its dispersing action) is inca- 

 pahie of further undergoing epipolic dispersion. 



In proof of this, the following experiments may be adduced. 



Exp. 1. A glass jar being filled with a quiniferous solution'*, a piece of plate-glass 

 was immersed in it vertically, so as to be entirely covered and to present one face 

 directly to the incident light. In this situation, when viewed by an eye almost per- 

 pendicularly over it, so as to graze either surface very obliquely, neither the anterior 

 nor posterior face showed the slightest trace of epipolic colour. Now the light, at 

 its egress from the immersed glass, entered the liquid under precisely the same cir- 

 cumstances as that which, when traversing the anterior surface of the glass jar, un- 

 derwent epipolic dispersion on first entering the liquid. It had therefore lost a pro- 

 perty which it originally possessed, and could not therefore be considered, qualita- 

 tively, the same light. 



Exp. 2. The epipolic tint is developed only on the surface of incidence. When 

 the solution is exposed to light in a glass vessel, the posterior surface, whether viewed 

 internally or externally, is quite colourless. Here again, since ingress and egress 

 into and out of a medium are, optically speaking, convertible, a qualitative analysis 

 at the surface of incidence would appear to be indicated. 



Exp. 3. A test cylinder filled to the height of two or three inches with the so- 

 lution was set upright on black velvet, its bottom being also shaded to the depth of 

 half an inch (to prevent reflected light from the bottom from reaching the eye). 

 The epipolic tint being now fully developed, a hollow parallelopiped of plate glass, 

 filled with the same solution, was interposed between the test cylinder and the inci- 

 dent light, side light being at the same time obstructed by screens duly placed. 

 Immediately the epipolic colour in the interior of the cylinder vanished altogether. 

 The transmitting vessel was now emptied of its contents and filled with pure water, 



* The solution here used and subsequently referred to (except when otherwise expressed) is formed by add- 

 ing to sulphuric acid, diluted to such an extent as just to bear being swallowed without pain, about one two- 

 hundredth part of its weight, (the weight, i. e. of the diluted acid) of sulphate of quinine. When of this strength 

 it is difficult to believe that a bottle half-filled with it contains a colourless liquid. When shaken, it tinges 

 the glass vividly blue : lively blue gleams are reflected from the interior, and from the capillary ring at the 

 surface level, &c. I may mention that in one instance a rose-coloured solution was formed, which I have 

 never been able to reproduce. The ingredients were taken from the very same parcels which gave the usual 

 colourless solution, and the mixture made in the identical vessel which had just recently served for the same 

 experiments, and which had not even been washed, and from which a colourless solution had just been emptied. 

 If owing to any foreign ingredient accidentally present, the quantity must have been inconceivably minute. 



