ON LIGHT. 355 



th.it in the plane of the section of the prism at right 

 angles to the edge. If, instead of cutting the crystal 

 into such a prism, it be formed into a flat plate, with its 

 faces parallel to the axis of the rhomboid ; such plate 

 will in like manner extinguish one of the pencils into 

 which a ray incident perpendicularly on it is divided, 

 allowing the other to pass j and the pencil so transmitted 

 is completely polarized in a plane perpendicular to the 

 axis of the plate. This property of a tourmaline plate 

 renders it invaluable as an optical instrument, affording 

 the readiest and most convenient means of procuring a 

 polarized beam of hght for the examination of crystals 

 and other purposes. Its only drawback is, that this 

 mineral is most commonly coloured with a strong tint of 

 blue or green, which affects the colour of the trans- 

 mitted light. Some specimens, however, while equally 

 effective in destroying one of the refracted pencils, are 

 yet but slightly tinged with colour as respects the other, 

 which is therefore transmitted fully polarized, but with 

 only a slight tinge of brownish yellow. The other 

 substance, of late much resorted to for the same pur- 

 pose, the iodo-sulphate of quinine, crystallizes in very 

 thin scales like mica, of a purplish-brown hue, which in 

 like manner polarize completely one half the incident 

 light, which passes freely through them ; the other half 

 being extinguished. This curious property was dis- 

 covered by Mr Herapath, who first formed the com- 

 pound in question. 



(136.) When two parallel plates of tourmaline cut 

 from the same crystal in the mode above described, or 



