Astronomical and Nautical Collections* 1§^ 



stalllzed plate : and this last case is the best fitted for om* 

 purpose, the others affording only phenomena which are pre- 

 cisely similar. 



In order to divide the light into two pencils which inter- 

 sect each other in a small angle, and which are thus fitted to 

 afford fringes, the apparatus of two mirrors is in general 

 preferable to the screen with two slits, because it affords 

 more brilliant fringes ; besides, it has here the advantage of 

 giving to the pencils the previous polarisation required for the; 

 experiment. It is sufficient for this purpose that the mirrors 

 should be of unsilvered glass, and inclined about 35° to the 

 incident j'ays: and care must be taken to blacken them on the 

 back, in order to destroy the second reflexion. We place near 

 them, in the path of the light, and perpendicularly to its 

 direction, a plate of sulfate of lime, or of rock crystal, cut 

 parallel to the axis, and a tenth or twentieth of an inch in 

 thickness ; inclining the principal section in an angle of 45^ 

 to the plane of primitive polarisation, which we have sup- 

 posed to be horizontal. The apparatus being thus arranged, 

 we shall only see a single group of fringes through the plates 

 as before its interposition, and occupying the same situation. 

 But if we put before the lens a pile of glass, inclined either 

 to the horizontal or to the vertical direction, we shall dis- 

 cover, on each side of the central group, another group of 

 fringes, which will be so much the more remote from it as the 

 crystallized plate is thicker. If we substitute for the pile of 

 glass a rhomboid of c^lcarious spar, of which the principal 

 section is in a horizontal or a vertical direction, we observe, 

 in each of the two images that it produces, the two additional 

 systems of fringes which had been before formed by the in- 

 terposition of the pile ; and it is remarkable, that these two 

 images are complementary to one another ; that is to say, 

 that the dark stripes of the one correspond to the bright 

 stripes of the other. 



We see in this experiment a new confirmation of the prin- 

 ciples which are demonstrated by the foregoing. The rays, 

 which have undergone the contrary refractions, cannot affect 

 each other ; because, when they emerge from the same 

 plate, in the case that we are considering at present, they 



