1830.] 
On the Polarization of Light. 
109 
this is a rule with all crystals, so that a method is hereby obtained for determin- 
ing whether the double refraction of a substance be attractive or repulsive. 
14. The tourmaline demonstrates, moreover, that the greatest portion of Iio-ht 
which escapes in all directions from any body by irregular reflection, has a contrary 
polarization to that portion which is regularly* reflected. This experiment should 
be performed in a dark room, by admitting a solar ray upon any substance 
(figure 19.) » 
15. Agate resembles tourmaline .somewhat in its action upon light. Brewster 
discovered that, when thin slices were cut perpendicular to the natural strata, they 
suffered all light indiscriminately to pass, but gradually as the stone became thick- 
er* ah the light was absorbed, but such as was polarized in the direction of the 
laminae. In pure tourmaline, no laminae can be perceived, but their existence is 
rendered evident by this same optical phenomenon. It resembles what was describ- 
ed in section eight, of the action of a number of glass plates. 
16. In sulphate of lime or gypsum, the axis of double refraction forms au 
angle of 16° 13', with the shortest side of its paralielogramic crystals, being also 
at right angles to the axis of crystallization, as are the laminae of the crystal, 
which may be cut off to any degree of thinness. 
17. The facility of procuring extremely thin plates of gypsum, led Arago to the 
investigation of a new species or modification, which he called “ moveable polariza- 
tion ”, because in its production, the atoms of light are supposed to be in a state of 
oscillation. The general fact is, (and it holds good for thin plates of all other 
crystals,) that when such a lamina is interposed before the second glass of the po- 
larizing instrument, (supposed to be fixed at 90°, or in a position for transmitting 
the ray of the tube,) then should the crystalline axis of the lamina be brought pa- 
rallel to the first plane of reflection, no portion of light will be diverted from its 
previous course, but should the lamina be turned in azimuth (figure 20), a portion 
of light will instantly be abstracted from transmission, and be reflected from the 
glass. This effect has been improperly termed depolarization , for it appears from 
examination by the tourmaline or other tests, that the reflected light has not ac- 
quired fixed polarization at right angles to that transmitted, but that it carries its 
axis of polarization in an azimuth double of the deviation of the axis of crystalliza- 
tion of the lamina, from the plane of first reflection or zero, figure 22. 
* 
■m 
fia 2 J 
