PREFACE. IX 



pression or dilatation, or its elasticity of volume, while B mea- 

 sures its resistance to distortion, or its rigidity. The equilibrium 

 of the medium, it may be shewn, cannot be stable, unless both 

 of these quantities are positive*. A Supplement to this paper 

 supplying certain omissions, immediately follows it. 



In the next paper, " On the Propagation of Light in Crys- 

 talline Media," the principle of Conservation of Work is again 

 assumed as a starting-point and applied to a medium of any 

 description. It is first assumed that the medium is symmetrical 

 with respect to three planes at right angles to one another, by 

 which supposition the twenty-one coefficients previously men- 

 tioned are reduced to nine. Fresnel's supposition, that the 

 vibrations affecting the eye are accurately in front of the wave, 

 is then introduced, and a complete explanation of the phe- 

 nomena of polarization is shewn to follow, on the hypothesis 

 that the vibrations constituting a plane-polarized ray are in 

 the plane of polarization. The hypothesis adopted in the 

 former paper that these vibrations are perpendicular to the 

 plane of polarization is then resumed, and an explanation 

 arrived at, by the aid of a subsidiary assumption unfortunately 

 not of the same simple character as those previously intro- 

 duced that for the three principal waves the wave-velocity 

 depends on the direction of the disturbance only, and is in- 

 dependent of the position of the wave's front. The paper 

 concludes by taking the case of a perfectly general medium, 

 and it is shewn that Fresnel's supposition of the vibrations 

 being accurately in the wave-front, gives rise to fourteen re- 



* In comparing Green's paper with the passage in Thomson and Tait's 

 Natural Philosophy above referred to, it should be remarked that the A of the 

 former is equal to the m - \n of the latter, and that B=n. 



