410 Sir D. Brewster on the Decomposition and Dispersion 



stood for some days. Tlie same property is possessed by almost 

 all the oils, in some of which the dispersed light is exceedingly 

 beautiful, varying from a pale green to a blue tint. 



The polarization of the dispersed beam in one plane, namely 

 in the plane of reflexion, is exhibited in several fluids and solids. 

 It is very marked in the bile of the ox, which disperses an olive- 

 green light; in a solution of gum-myrrh in alcohol diluted 

 with water, which disperses a bright white beam ; and in an 

 orange-coloured glass, which disperses a pale greenish beam. 



In many fluid solutions, the beam with a qnaquaversus po- 

 larization is very intense, when compared with the faint pencil 

 which is polarized in the plane of reflexion ; but in a specimen 

 oi yellow Bohemian glass, which gives a copious and l)rilliant 

 gree7i beam by dispersion, the whole of the beam possesses a 

 qnaquaversus polarization. 



When we view the dispersed beam in different azimuths, 

 some very interesting phasnomena present themselves to our 

 notice. In general, the colour of the dispersed light suffers a 

 considerable change, passing, between the azimuths of 90° and 

 180°, from the colour of the dispersed beam to the colour of 

 the transmitted beam. This effect is finely seen in the alco- 

 holic solution of tea, where the brilliant red light passes into 

 an olive tint; but it is still more remarkable in a mixture of 

 Prussian blue and water. The dispersed beam is polarized in 

 the plane of reflexion. It is bluish in the azimuth of 90°; 

 pinkish about the azimuth of 100°; greenish in that of 120°; 

 bluish \n a7Am\\\\\ 150"; and again pinkish in azimuth 170°. 

 These three last tints may be all seen at the same time. 



Such are the general phsenomena of internal dispersion, a 

 subject which promises to throw some light on the constitution 

 of those solid and fluid bodies by which it is produced. The 

 apparently supetjicial dispersion in the quinine solution, to 

 which Sir John Herschel has given the name of cpipolism, is 

 obviously a single case of the general phcenomenon, in which 

 the ordinates of the curve of dispersion diminish rapidly after 

 the light has entered the stratum nearest the surface; while 

 the real epipolism, which he ascribes to fluor-spar, so far from 

 being an action of the surface, is much less so than that of the 

 quiniferous solution, and entirely similar in its character to 

 that which is produced by the fluids and solids which I have 

 examined. 



The phasnomenon of internal dispersion, when considered 

 merely as a case of reflexion and polarization, possesses much 

 novelty and interest. If the exciting beam, as we may call it, 

 is cylindrical, we have before us an experiment, in which the 

 phaenomena of cylindrical rejiexion and cylindrical polariza- 



