732 



Mr. A. E. Garrett on Electrical 



The production of fogs would lead us to think that the 

 ions are bodies o£ considerable size ; in the case of the positive 

 ions other evidence points to the same conclusion. 



II. Variation of the Leak with Changes of Temperature. 



_b_ 



Richardson has found * that the formula l — a6h e can be 

 used to express the relation between the saturation-current I 

 and the absolute temperature in the case of a hot plaiinum 

 wire and a metal cylinder surrounding it in a high vacuum, 



I find in all the cases tried that a similar formula repre- 

 sents the relation between I and 6. 



If logs, are taken of both sides of the equation, since a and 



b are constants it will be seen that if -^ is plotted for abscissae 



and Jlog e — log J for ordinates, the result obtained should 

 be represented by a straight line. Fio-. 2. 



These earlier results were ob- ro£i£cr/?oM£T^ 

 tained with voltages which were 

 too small to produce saturation- 

 currents, although they were suffi- 

 ciently large for the rate of leak 

 to depend rather upon the total 

 number of ions present than their 

 velocity. In order therefore to 

 ascertain whether Richardson's 

 formula held for results obtained 

 with sufficient voltage to produce 

 a saturation-current, some zinc 

 iodide was heated up and tested in 

 apparatus in which saturation could 

 be obtained. This is shown in 

 fig. 2. 



A is a brass tube earthed. B and 

 C are brass plates, the material to 

 be tested being placed on the lower 

 of the two plates ; the plates were 

 separated by a few millimetres. 

 D is a battery of storage-cells. E a 

 glass tube fitted with corks and / 

 made air-tight. To &*™ 



The following results were obtained by this means for 

 negative carriers : — 



* Proc. Camb. Phil. Soc. xi. p. 286 (1902). 



