﻿S82 Dr. E. N. da C. Andrade on the Carriers of 



E. Marx* has also measured the velocity of the positive 

 carriers ; he assumes, however, for the purpose of simplifying 

 his calculations, that the whole ionization takes place in the 

 body of the vapour, which is certainly incorrect when 

 electrodes oE glowing platinum stand in the metallic vapour, 

 as in his experiments (cf. p. 874). His work is of great 

 interest, but it is not adapted to give a correct value of the 

 velocity. 



When this work was practically completed a paper by 

 Lusby t on the subject came to my notice, and a second 

 paper by this author, including his previous results, has 

 just been published J. He brings forward criticism essen- 

 tially agreeing with that already given here against the 

 previous work. He has made measurements with Wilson's 

 original method, but has measured the gradient of potential 

 in the flame from one electrode to the other, with the salt 

 vapour present in the flame. At normal flame temperature 

 the gradient in the vapour is very small, that in the free 

 flame large : Lusby takes the gradient in the vapour as the 

 one relevant, which seems to be erroneous as already 



cm. 



explained. With this he gets a value of 290 '- As he 



1 ^ sec. 



has given the curve of potential for one case in his paper §, 

 we can obtain the correct velocity according to the views 

 here advanced. Taking the gradient in the flame just be- 

 neath the salt (50 ), we get as the value of the velocity 



cm 

 4*1 : , which agrees pretty well with our average value 



of 2-5^- 



sec. 



His assumptions as to the gradient in question will account 

 for a peculiar effect noticed by Lusby. He has made 

 measurements at different temperatures, and finds that down 

 to 1450° absolute the velocity is roughly proportional to 

 ^/6, but that below this the values of the velocity are very 

 much smaller than those given by ^/6. Now down to 

 1450° abs. the distribution of potential varies very little 

 with the temperature, but below this the gradient in the 

 vapour increases very much, and this is the main cause why 



* E. Marx, Ann. der Physik [4] ii. p. 768 (1900). 

 t S. G. Lusbv, Proc. Cambridge Phil. Soe. vol. xxi. cart 1, p. 26 

 (1911). 



t S. G. Lusby, Phil. Mag. Not. 1911, p. 775. 



§ Proceedings Cambridge Phil. Soc. loc. clt. p. 30. 



