410 Mr. R. T. Glazebrook on the Molecular 



the electro-kinetic is 



being obtained from tbe kinetic by substituting —- for p, and 



p, q, r for £, ?;, £. If, then, in the kinetic energy we have the 

 term 



in the electrokinetic energy we have 

 and this, on integrating by parts, gives 



which is Mr. Fitzgerald's additional term. 



In fact, if we consider the medium as consisting of a num- 

 ber of separate molecules, the kinetic energy has its ordinary 

 meaning ; the electrokinetic is that part of the kinetic energy 

 which depends only on the rotation of the molecules ; for this 

 rotation alone produces magnetic force. If a transverse wave 

 of disturbance travels through the medium, the translational 

 motion of each molecule becomes known to us as light, while 

 to the rotational velocity we give the name of magnetic force. 

 Since the direction of rotation at each point is reversed many 

 times a second, we cannot produce magnetic force by a wave 

 of light. If, however, magnetic force exists in the medium 

 independently of the light, the translational motion of the 

 molecules is modified thereby. 



Now let us consider a molecule of the medium, which we 

 shall suppose moves as a rigid body, with angular velocities 

 a>ij co 2 , ft> 3 . If £, 77, f are the coordinates of the centre of 

 gravity, and dx dy dz the volume, A, B, C the radii of gyra- 

 tion about axes parallel to those of coordinates, which we 

 assume to be principal axes, then the kinetic energy of this 

 molecule is 



ip& + V 2 + t) dx dy dz + p(A 2 col + B 2 ^ + C 2 a^) dx dy dz. 



In a material medium, A, B, C being proportional to the linear 

 dimensions of the molecule, the last term, A 2 a>j + B 2 ^ + C 2 ^ 

 vanishes compared with the other, the molecule being very 



