220 FRESNEL. 



There must be then, in each of these rays, north and 

 south sides in some way different from their east and 

 west sides. And further, the north and south sides of 

 the ordinary ray ought to have precisely the same prop 

 erties as the east and west sides of the extraordinary ray ; 

 so that if this last ray make a quarter of a circuit about 

 the line of its length it will be impossible to distinguish 

 one from the other. The rays of light are so subtle that 

 thousands of millions of these rays can pass simultane 

 ously through the eye of a needle without interfering : 

 yet we find ourselves obliged to take into account the 

 idea of their sides, and to recognize, on their opposite 

 sides, dissimilar properties. 



When speaking of a magnet, natural or artificial phys 

 icists affirm it to have poles. They mean only that cer 

 tain points on its surface are found endowed with certain 

 properties which are not found, or at least only show 

 themselves feebly, at any other points. We have, then, 

 equal reason to say the same thing of the ordinary and 

 extraordinary rays of light which proceed from the divis 

 ion of the beam which passes through Iceland spar ; and 

 in contradistinction to the natural rays in which all points 

 appear alike, we may rightly call them polarized rays. 



In order, however, that we may not extend beyond its 

 proper limits the analogy of a polarized ray and a mag 

 net, it is important to remark well that in the ray, the 

 poles diametrically opposite appear to possess exactly the 

 same properties ; whilst the dissimilar poles are situated 

 on sides of the ray whose positions are at right angles to 

 each other. 



The lines resembling diameters, which join the similar 

 poles, in every ray deserve particular attention. When 

 ever, in two distinct rays, these lines are parallel, we say 



