DOUBLE REFRACTION. 151 
Thus, then, the two rays on emerging from the first 
crystal, instead of being changed in their nature, are ex- 
actly alike; it suffices, to show them undistinguishable 
from each other, to make one of these rays turn round 
the line of its own direction through 90°. Thus we are 
brought by the phenomena of double refraction to distin- 
guish in rays of light different sides endowed with dif- 
ferent properties. We are brought by observation to 
acknowledge that the extraordinary ray emerging from a 
rhomb revolve continuously, we have four images (as at c) unequally 
bright by pairs: at 45° four equally bright (p), the other pair now be- 
come faint (£), until at 90° they are reduced to two, (F). The same 
changes are repeated at G and H; when at 185° the four are equally 
bright, till, after two become faint at 1, we arrive at 180°, where, as 
at K, the two brighter coalesce into one. The same changes then 
take place in reverse order; four images at 225°, two at 270°, four at 
815°, and lastly two at 360°. 
To give any idea of the analysis of these phenomena, it is necessary 
in the first instance to observe accurately the form of the crystal, and 
obtain a distinct idea of the terms the azis and the principal section of 
the crystals, which will be understood at once by the aid of the an- 
nexed diagram; where taking the short diagonal of two of the oppo- 
site faces of the crystal, as A D, the plane passing through it A p x, is 
the principal section, and the diagonal of that plane a x, the azis of 
the crystal, The double refraction of a ray R is represented by its 
division into two rays, 0 the ordinary, and E the extraordinary. 
Whatever theory we adopt as to the nature of light, the phenomena 
can only be explained by supposing a section of each of the rays 
