128 On the Structure of doubly refracting Crystals. 



laminae, and to the direction of the serpentine lines, give addi- 

 tional probability to this conclusion. 



Here, then, we have a case of the most unequivocal kind, in 

 wliich one image of a doubly refracting crystal is produced by 

 one structure, or by one part of the crystal, while the other 

 image is produced by another structure, or another part of the 

 crystal; and hence we are led to conclude, in general, tkat the 

 two images exhibited by all doubly refracting bodies, are forn)cd 

 by two different structures, related to some axis or fixed line in 

 the primitive crystal. - Whether this difference of structure i:s 

 produced by a difference in the arrangement of the elementary 

 molecules, or is owing to a combination of different ingredients, 

 is a point which still remains to be determined. 



The phainomena presented by the agate and the carbo!iate of 

 barytes, convey still further information resjjecting the structure 

 of these imperfect crystals. In one direction, the light trans- 

 mitted by the agate is wholly nebulous; the perfect image being 

 converted into a shapeless cloudy mass, and confounded with 

 the nebulous image. In another direction, one of the images is 

 distinct and perfectly formed ; and in one specimen, which has 

 the faculty of depolarization, there nmst necessarily be two per- 

 fect images. In a prism of the carbonate of barytes, both the 

 images .were imperfect. In a second prism, the one image was 

 nebulous, and the other distinctly formed ; while, in other prisms, 

 there was a rapid approximation to two perfect images. Hence 

 it follows, that the imperfect structure, which, in general, trans- 

 mits only a mass of nebulous light, allows a distinct image to be 

 formed, when the rays are incident in one particular direction ; 

 while the perfect structm-e, which in general gives a distinct 

 image, allows an imperfect image to be formed, when the light 

 penetrates it by a particular ])at;i. 



These inferences, which I conceive to t)e irresistible, have a 

 higher degree of importance than we may at first be disposed to 

 attach to them. They form a real step in the explanation of 

 double images, and indicate a part of that structure which is 

 necessary to their formation. The other phajnomena of double 

 refraction are still involved in o])scurity. I'he opposite polariza- 

 tion of the two pencils may I)e explained by supposing the cry- 

 stal to consist ol laminaj inclined in various directions ; and, as 

 I liave shown in another place *, the same phaenomena may be 

 actually produced by an artificial crystal composed of bundles of 

 tjlass plates. The most perplexing point, however, is the extra- 

 oidinarv refraction which takes place at a perpendicular inci- 

 dence. ^\'hethcI this phoenomenon is the result of an extraorr 



* riiil. Trans. Loiul. J814, p;irt i. p. 1'AO. 



dinary 



