358 ON LIGHT. 



the celestial concave, which is 90 remote from the sun's 

 place, the amount of polarized light which it sends to 

 the eye bears a very considerable proportion to the total 

 illumination, amounting to nearly a fourth of the whole. 

 At every other inclination of the visual ray to the direct 

 sunbeam, the proportion is less, and in the neighbour- 

 hood of the sun, or of the point on the horizon directly 

 opposed to it very small. When examined in a mode 

 hereafter to be described, by the intervention of a tour- 

 maline plate, and a crystallized lamina, this gives rise to 

 a series of exceedingly beautiful and brilliant phsenomena, 

 productive of the greatest astonishment to those who 

 learn for the first time by their exhibition, that totally 

 different, and even opposite qualities, characterize dif- 

 ferent portions of an illuminated surface apparently so 

 perfectly uniform and homogeneous. This effect is sup- 

 posed to originate in the reflexion of the solar rays on 

 the particles of the air itself, an explanation encumbered 

 with many difficulties, but the best (indeed the only one) 

 that has yet been offered. 



(139.) Polarization interpreted on the undulatory theory. 

 According to any conception we can form of an elastic 

 medium, its particles must be conceived free to move 

 /'within certain limits greater or less according to the 

 coercive forces which may restrain them) in every direc- 

 tion from their positions of rest, or equilibrium. It by 

 no means follows, however, that the nerves of the retina 

 are equally susceptible of excitement by vibrations of the 

 luminiferous ether (in which they may be conceived im- 

 mersed) in all diiections. In the case of sound, the 



