394 Astronomical and Nautical Collections. 



rallel to the primitive plane, and those which answer to an 

 uneven number are polarised at the azimuth 2 i, and that the 

 rest exhibit only a partial polarisation. 



The experiment with the two rhomboids affords us a sin- 

 gular example of rays polarised in two orthogonal planes, 

 which produce, by their union, light completely polarised in 

 an intermediate direction : and this circumstance supports 

 the hypothesis which has been mentioned in speaking of the 

 law of Malus, when it was hinted that the vibration of light 

 might be executed in planes parallel or perpendicular to 

 that of the polarisation. 



Phenomena of the same kind are afforded, in the same cir- 

 cumstances, by thin crystallized plates ; when the rays have 

 been polarised in a common plane, before their entrance into 

 the plates, and when the difference of the paths of the two 

 systems of undulations, at their emersion, is equal to a whole 

 number of semiundulations : when it is an even number, the 

 whole light that emerges is polarised in the primitive plane, 

 and when an odd number, it is polarised at the azimuth 2 i. 

 For example, if the angle i is 45°, that is, if the axis of 

 the plate makes an angle of 45° with the primitive plane, 

 the whole light will be polarised in the first case in the 

 direction of the primitive plane, at 45° from the axis, and in 

 the second case, at the azimuth of 90°, or perpendicularly to 

 the primitive plane. But it does not follow, because the 

 whole compound light is thus polarised, that the ordinary 

 and extraordinary rays composing it are polarised in the 

 same direction, as we have seen by the experiment with the 

 two rhomboids. In fact, the circumstances of the experi- 

 ment are the same, the only difference is that the two 

 systems of undulations transmitted by the crystallized plate 

 are parallel to each other, while those which pass through 

 the rhomboids form a sensible angle, so that they are not 

 visible without the employment of a luminous point, and a 

 lens, which are required to make the effects of their inter- 

 ference perceptible. But, in consequence of this inclination, 

 they exhibit at once all the differences of routes in the dif- 

 ferent points of the groups of fringes which they produce, 

 and collect in this manner, into a single picture, all the cases 

 afforded by crystallized plates of different thicknesses. 



