284 SPECIAL RESULTS IN THE UEANOLOGICAL 



reflected or refracted, whether it emanates from a solid, from 

 a liquid, or from a gaseous body, and even announces its 

 degree of intensity/' (Kosmos, Bd. i. S. 35, Bd. ii. S. 

 870; English Ed., Yol. i. p. 37, and Tol. ii. p. 329.) It 

 is essential to distinguish between natural light as it pro- 

 ceeds directly from the sun, the fixed stars, or gas-flames, and 

 is polarized by reflection from a glass plate under an angle 

 of 35 25', and the polarized light which radiates spon- 

 taneously as such from certain substances (glowing solids as 

 well as liquids). The polarised light given out by the last- 

 mentioned class of bodies proceeds very probably from 

 their interior ; on passing from a denser body into the 

 thinner surrounding atmospheric strata, it is refracted at 

 the surface, and a part of the refracted ray returns inwards 

 and becomes polarized by reflection, while the other portion 

 presents the properties of light polarised by refraction. 

 The chromatic polariscope distinguishes between these two 

 kinds of light by the opposite position of the coloured com- 

 plementary images. Arago has shewn by careful experi- 

 ments extending back to before 1820, that radiant solid 

 bodies, e. g. } a red-hot iron ball, or glowing, shining, 

 molten metal in a liquid state, give out simply natural light 

 in the rays which issue from them in a perpendicular direc- 

 tion, whereas the luminous rays which arrive at our eyes 

 under very small angles from their edges are polarized. 

 If we now turn the polariscope, by which these two kinds 

 of light are distinguished from each other, to gas-flames, 

 no polarisation is discovered, however small may be the 

 angles at which the rays emanate. Although light may 

 be produced in the interior of the gaseous body, yet, in 



