Were tourmaline like other doubly refracting bodies, each of the transmitted rays would be double; but that mineral, when of a certain thickness, after separating the light into two polarized pencils, absorbs that which undergoes ordinary refraction, and consequently shows only one image of an object. On this account tourmaline is peculiarly fitted for analyzing polarized light, which shows nothing remarkable till viewed through it or something equivalent.
The pencils of light, on leaving a double refracting substance, are parallel; and it is clear, from the preceding experiments, that they are polarized in planes at right angles to each other (N. 206). But that will be better understood by considering the change produced in common light by the action of the polarizing body. It has been shown that the undulations of ether, which produce the sensation of common light, are performed in every possible plane, at right angles to the direction in which the ray is moving. But the case is very different after the ray has passed through a doubly refracting substance, like Iceland spar. The light then proceeds in two parallel pencils, whose undulations are still indeed transverse to the direction of the rays, but they are accomplished in planes at right angles to one another, analogous to two parallel stretched cords, one of which performs its undulations only in a horizontal plane, and the other in a vertical or upright plane (N. 206). Thus the polarizing action of Iceland spar and of all doubly refracting substances is to separate a ray of common light, whose waves or undulations are in every plane, into two parallel rays, whose waves or undulations lie in planes at right angles to each other. By a simple mechanical law each vibratory motion of the first is resolved into two vibratory motions at right angles to one another. The ray of common light may be assimilated to a round rod, whereas the two polarized rays are like two parallel long flat rulers, one of which is laid horizontally on its broad surface, and the other horizontally on its edge. The alternate transmission and obstruction of one of these flattened beams by the tourmaline is similar to the facility with which a card may be passed between the bars of a grating or wires of a cage, if presented edgeways, and the impossibility of its passing in a transverse direction.