Asymmetric Emission of Secondary Rays. 149 



Since the ratio of the number of ions emitted forward to 

 the number emitted backward is 



-Jr= I n dx-r- \ n dx 

 Y J * J-i x 



(3) 



we find from (6) ^ = 7, and from (7) i/r = co . 



The possibility of (6) appears to be eliminated by the 

 results of Bragg and Mad sen's * experiments on hard y rays 

 which, in the case of carbon, gave a ratio of about 20 to 1. 

 Nevertheless it cannot be said to have been demonstrated 

 that the electromagnetic theory is incapable of accounting 

 for these effects. 



In the neighbourhood of the visible spectrum, let us say 

 when i/ = 10 15 sec. -1 the ratio of u to v deduced from 

 equation (2) is only about 1 to 500. Thus even the compa- 

 ratively small differences between the emergence and incidence 

 emission found by Stuhlmann f seem (arger than would be 

 expected. It is probable, however, that somewhat different 

 considerations enter in these cases. The value of v is now com- 

 parable with the value which it is necessary for an electron to 

 have in order to escape from the metal at all. A small differ- 

 ence in the value of v 2 in favour of the direction of emission 

 will be equivalent, as compared with the unfavourable case, 

 to an extension of the illumination towards the ultra-violet. 

 The equivalent extension Bv, of the range of frequency of the 

 illumination, will be proportional to the difference in v 2 , and 

 as the range of v which is effective in causing emission is 

 only a fraction of the total range from to v the effect will 

 be correspondingly multiplied. A difference in the number, 

 with v constant, will act in the same way as a difference in 

 v, but not quite so directly. On the above view, with the 

 same illumination, the difference will tend to be greater 

 when the effective range of the illumination is shorter. This 

 is the case with the more electronegative elements, as the 

 minimum frequency and velocity required for emission are 

 greater for them. Although, no doubt, there are other 

 factors entering when illumination from a source like the 

 iron arc is used, it is interesting to observe that, of the 

 differences recorded by Stuhlmann, the greatest is in the 

 case of the most electronegative element, platinum, and the 

 least in the case of the most electropositive element, 



magnesium. 



* Trans. R. S. of S. Australia, vol. xxxii. p. 1 (1908) ; Phil. Mag. 

 1909. 

 t Phil. Mag. vol. xxii. p. 863 (19] 1). 



