114 CONDUCTION AND RADIATION [CH. x. 



of the fact and of its elementary theory, is that given 

 by the present writer in two articles in the 

 Electrician for February and March, 1897,* which 

 are worth referring to as representing incipient ideas 

 on the subject before the full significance was grasped. 

 The high value of the e/m ratio, viz., ^x 10 7 c.g.s., 

 or fifty million coulombs per gramme, instead of the 

 moderate electrolytic value, is spoken of on page 643 

 as a difficulty ; and a FitzGerald suggestion amounting 

 virtually to the beginnings of an electron theory of the 

 Zeeman effect is hinted at. Likewise an extremely 

 short way of expressing the theory of the motion is 

 given by the writer, in the following form : 



Consider the resolved part of any orbital motion 

 projected on to a plane normal to the applied magnetic 

 field H ; and let the angular velocity be o>, at any 

 point of an orbit where the radius of curvature is 

 r ; then the field will exert a radial component 



which will represent an increment or decrement of 



centripetal force -, / 2 \ 



a (mrur) ; 



whence it follows, to a first approximation of order of 

 magnitude, that TT 



doo= - , 



2m ' 



and the change of frequency caused by the magnet- 

 isation in the transverse components of the radiation 

 will therefore be 



4-Trm' 



The other or longitudinal component of the original 

 orbit will manifestly be unchanged. This is far from 



* See Lodge, Electrician, vol. 38, pp. 568 and 643. 



