300 jDeterm'malion of I he Laws according to which 



reflection two distinct effects. It impresses in the first place on 

 a part of the incident light the polarization moveable around 

 the plane of incidence ; that is to say, it causes to oscillate the 

 luminous molecules on both sides of this plane, in Uie same way 

 as a thin crystallized plate or one of weak polarizing power 

 makes them oscillate on both sides of its principal section ; and 

 in both cases equally the tints puss through the wliole series of 

 reflected and transmitted rings of Newton. Rut besides, the 

 metallic surface imprints on a white portion of the incident light, 

 the fixed polarization in the plane of incidence, in the same way 

 as a thick crystallized lamina, or one of energetic polarizing 

 force, gives to the light which passes through it the fixed po- 

 larization, in two rectangular directions ; and in the same way 

 as in all crystallized bodies I have shown that the luminous 

 molecules pass progressively from the moveable to the fixed po- 

 larization, when they have penetrated to a certain depth, in the 

 same way in every reflection between metallic lamina we ob- 

 serve that a part of the light which had undergone the move- 

 able polarization in the preceding reflections, takes the fixed 

 polarization, which it can never afterwards quit if the reflecting 

 lamince are parallel ; so that in this case, after a number of re- 

 flections, more or less considerable, according to the nature of 

 the metal aiul that of the polish which was given to it, we ought 

 to find, and in fact we do find, almost the whole light polarized 

 fixed according to the plane of reflection. In the reflection upon 

 steel, and probably on the other metals which take a very lively 

 specular polish, the portion of white light which is thus taken 

 off from the moveable polarization is incomparablv the strongest; 

 so that the phoenomenon of the colours wViich the moveable po- 

 larization can alone produce becomes insensible, or cannot be 

 perceived but in certain peculiar po-sitions vvhich theory alone 

 can point out. Thus Mr. Brewster has announced that this 

 phsenomenon had not taken place upon steel and upon the allov 

 vvhich is used for mirrors ; but by guiding myself by the indica- 

 tions of theory, I succeeded in observing it in a manner not to 

 be doubted, even upon the best polished steel. When we em- 

 ploy laminaa of silver which have received the specular polish, 

 the portion of the light which takes the fixed polarization at 

 each reflection is still more considerable ; but it is nevertheless 

 much less than upon the two metals vvhich I ha.ve cited ; by a 

 necessary compensation, the portion vvhich takes the moveable 

 polarization is gi eater, and the phsenomenon of the tints then 

 becomes finer and more casv to observe. But the direction of 

 polarization of the white fasciculus being precisely intermediary 

 between those of the coloured fasciculi, it results that it also 

 mixes with them in the refraction operated by the rhomboid ; 



and 



