[ 147 ] 



V. 'Aiii6p(j)cjTa, No. II. — On the Epipblic Dispersion of Light, heing a Supplement to a 

 paper entitled, " On a Case of Superficial Colour presented ly a homogeneous 

 liquid internally colourlessV By Sir J. F. W. Herschel, Bart., KM., F.R.S., §c. 



Received March 6, — Read April 3, 1845. 



In reasoning- on the peculiar coloured dispersion operated on a portion of a beam 

 of white light intromitted into a solution of sulphate of quinine, it occurred to me 

 as a subject well worthy of inquiry whether the rays so selected for dispersion 

 and thus singularly separated from the rest, were disting-uished by any other pecu- 

 liarity ; whether in effect an analysis of the incident light into two distinct species 

 qualitatively different had been performed, or merely a simple subdivision, such as 

 takes place, for instance, in partial reflexion, as in the phenomena of the colours of 

 thin plates. Another interesting subject of inquiry presents itself in the laws which 

 regulate this singular mode of dispersion itself, which, for brevity, I shall venture to 

 call (at least provisionally) epipblic, from eTriTroXi), a surface, the seat of the dispersion 

 being at or very near the intromitting surface. 



As regards the question of analysis, two modes of examination present them- 

 selves, viz. either, — I, by subjecting the dispersed portion of the light to experiment, 

 or, 2, the residual portion, which, having escaped dispersion, preserves the unity of its 

 direction ; and on that account, as well as by reason of its vastly superior intensity, 

 offers itself more readily to experimental inquiry. 



The colour of the dispersed portion being blue, that of the residual beam ought, 

 of course, to verge towards orange. But owing to the large excess of undecom- 

 posed white light present, this tendency is inappreciable ; and the regularly trans- 

 mitted beam is not to be distinguished by the eye from white light. Another reason 

 is, that some portion of the dispersed necessarily mingles with the regularly trans- 

 mitted beam, the medium being equally permeable to both ; so that in viewing an 

 extensive white surface (the cloudy sky for instance, or a piece of white paper), the 

 regularly transmitted ray reaching the eye in any given direction, that is, from any 

 one point in the luminous surface, has, intermingled with it, a dispersed ray from 

 every other point of that surface, the totality of which goes to restore to it some 

 material portion of the blue light which it lost by dispersion at its intromission. 



In the ordinary production of colour in liquids by absorption of the comple- 

 mentary tint, the smallest preference of one over the other coloured rays may be 

 magnified and brought into evidence as a cause of coloration by increasing the 

 thickness of the transmitting medium, or by passing the light successively through 



u2 



