366 Prof. Owens on Thorium Radiation. 



When the air in the vessel was not drawn through but was 

 simply agitated by a set of vanes attached to a vertical shaft 

 fitting tightly through the upper cover and operated by a small 

 motor, the side outlets being closed, the conduction-current 

 actually increased with the speed of the vanes, showing that 

 a mere motion of the air is not sufficient, but that it must be 

 removed from the vessel to decrease the value of the current. 

 While the experiments tried are not conclusive, they show 

 that the effect depends on the actual passage of gas through 

 the box, the thickness of the radiating layer, the nature of 

 the salt used, and the degree of its protection, and so indicate 

 that the cause, whatever it is, lies close to the surface of the 

 active material. It is possible that some intense type of 

 radiation coming from the body of a thick layer of certain 

 salts changes the nature of their surfaces, forming in the neigh- 

 bourhood a more active material which if removed from the 

 containing vessel diminishes the amount of ionization pro- 

 duced. 



All the observations recorded below were taken after the 

 conduction-current had come to a steady value. 



2. The Relation between Current and Electromotive Force. 

 ■ — The explanation of the general form of the curves showing 

 the relation between current and electromotive force in ionized 

 gases was pointed out by Thomson and Rutherford* in 1890. 

 For gas exposed to uranium radiation the relation was in- 

 vestigated the following year by Becquerel | and de Sinolan 

 and Beattie J, and again in 1899 by Rutherford §. 



Using potentials less than a volt with plates close together, 

 the latter found the conduction-current to vary with the sign 

 of the lower plate, and to have a certain small constant value 

 when no external electromotive force was acting, this being- 

 probable due to the contact-difference of potential between 

 the uranium salt and the plate upon which it rested, and also 

 to the different velocity with which the positive and negative 

 carriers diffuse. For large electromotive forces no appre- 

 ciable difference in value of the current was observed whether 

 the uranium was charged negatively or positively. 



In the March number of the Phil. Mag. for this year 

 Professor J. J. Thomson has given a general expression for 

 the relation of the current and potential-difference between 

 two parallel plates bounding a volume of ionized gas. His 

 final equation is of the form 



V=A*' 2 + B/, 



* Phil. Mag. Nov. 1896. f Comptes Rendus, pp. 438, 800 (1897). 



% Phil. Mag. p. 418 (1897). § Phil. Mag., Jan. 1899. 



