402 Lord Blythswood and Dr. Marchant on 



This gives a mean value 



^ = 306 x 10- 6 . ^w = 9-13 x 10- 13 , 

 H X 2 H 



\ as before, being expressed in 10~ 10 metres. 



It will be noticed that the value of 8\ 2 /H for the field 

 22,300 is somewhat low; but this may be explained by the 

 fact that this measurement was obtained by adjusting the 

 magnetic field until the right-hand component of the left- 

 hand order in the eyepiece was superposed on the left-hand 

 component of the right-hand order, an adjustment which it 

 is somewhat difficult to make accurately 



The values given above are not accurate to more than 

 2 per cent. 



It will be of interest to calculate the value of e/?n from 

 Larmor's equation (1), especially for the two yellow lines of 

 mercury, since these are produced by relatively simple ionic 

 movements. We have 



ni - n2= 2W' 



where e is the charge on an ion of mass m, 



H is the intensity of the magnetic field, 

 v the velocity of light in air, 



n x and n 2 the frequencies of the outer components of 

 the triplet ; 



hence vSX eH eH . . 



-^s- = zr 9 = « m electromagnetic units. 



For 



X = 5790, */m = 17-2xl0 6 , 



\=5758, e/m = 18'9xl0 6 , 



\=5460, <?/m=20'5xl0 6 , 



X=4358, ^/m = 28-0xl0 6 ? 



For the blue and green lines, these values are calculated 

 from the values of B\ 2 ', i- e. the mean dispersion between the 

 outer components of what was, under a weak field, a triplet. 



There is one point of interest with reference to the blue and 

 green lines which may be further noticed. The value of 

 S\/\ 2 H with the green line, is the same, both for the compo- 

 nents of the inner triplet and for the components of the doublet 

 into which the outer lines split. In the blue line (4358) 

 the value of S\/X 2 H for the components of the inner doublet 



* Phil. Mag. Dec. 1897, vol. xliv. p. 503. 



