380 ON LIGHT. 



would have been wholly extinguished at each quarter 

 revolution. 



(157.) Another mode of communicating circular polariz- 

 ation to a ray, is to transmit it at a perpendicular in- 

 cidence through a parallel plate of any perfectly colour- 

 less and transparent doubly refracting crystal, of such 

 thickness, that in the passage through it of the two 

 waves, parallel to its surfaces, into which the incident 

 wave (supposed plane) is divided, (the one conveyed by 

 ordinary refraction, the other by extraordinary, and there- 

 fore travelling with different velocities in the crystal,) 

 the one shall have gained or lost, after emergence, 

 exactly a quarter of an undulation on the other. For as 

 the corresponding rays emerge of equal intensity, and 

 oppositely polarized, they here also fulfil all the con- 

 ditions of circular polarization. If the thickness of the 

 plate be such, that the difference of phases is more or 

 less than an exact quarter (or any number of quarters) 

 of an undulation, the compound ray will be elliptically 

 polarized, and the degree of ellipticity will be determined 

 by the thickness of the plate. 



(158.) It may be asked, in what does a ray so circu- 

 larly polarized differ from an ordinary unpolarized ray, 

 seeing that the latter may always be regarded as com- 

 pounded of two ordinary rays of half the intensity oppo- 

 sitely polarized? We reply, in this: viz., that if again 

 transmitted through another such glass parallelepiped, 

 similarly situated, the difference of phase will be doubled. 

 The emergent ray then will consist of two equal rays 

 oppositely polarized (and therefore not interfering), dif- 



