on the Corbino Effect. . 701 



It is obvious that the simple theory of the Corbino effect 

 is not broad enough to cover the results we have obtained. 

 It seems probable that, as Richardson* has suggested, it is 

 necessary to take into account polarization effects. In the 

 Corbino effect these would enter in a double manner : first, 

 as in the Hall effect when measured by steady currents, in 

 altering the forces acting in the free electrons ; and second, 

 in contributing the polarization current to the current carried 

 by free electrons. The polarization current would be opposite 

 in sign to the free electron current, and thus the Corbino 

 effect receives a qualitative explanation. It is hoped that 

 further light may be thrown on this question by experiments 

 that are in progress with alternating currents and alternating 

 magnetic fields. 



It is of some interest to calculate the mean free path of 

 an electron in those metals in which the Corbino effect has 

 the sign predicted by the simple theory of free negative 

 electrons. The expression previously deduced for the circular 

 current is : 



~ HTI e , r 2 

 C= 7— . .log 



1-t m ° i\ 



where I is the radial current, H the magnetic force, T the 



p 

 free time of an electron. - = 1*7710". and r* and r x the 



n 



external and internal radii of the disk. Since 



C_ n 

 I ~M' 

 we may write 



m n 1 



1 ''2 



T = 4tt 



e HM, 



r\ 



M was found In* the method that has been described to be 

 5*4 . 10 4 . 



For copper H _. . 9 



SO 



that T = 2-8.10" 



The free times for the other metals may be found from 



this by multiplying by the numbers in the last table which 



give the ratio of the Corbino effects. The mean free path is 



found by multiplying the free time by the mean velocity 



* Phil. Mag. April 1912, p. 614. 



