Electrification at Liquid- Gas Surfaces. 303 



practically the same velocity as air. For hydrogen an 

 average of three sets of readings gave 4*2 . 10~ 4 ; of three 

 others 4*1. 10~ 4 , From Professor Thomson's experiments 

 on the electrification of drops one might expect to find a 

 difference in the velocity of hydrogen and oxygen. In his 

 experiments the water was carefully freed from dissolved 

 gases and a perfectly fresh surface was exposed with each 

 drop, conditions difficult to obtain in a rotating cell. 

 However, water thoroughly boiled was tried but no change 

 of sign occurred with hydrogen as compared with oxygen, 

 and, as far as could be observed, there was no difference in 

 velocity. No actual measurements were made on account of 

 the rapidity with which the bubble was absorbed by the 

 water. 



Experiments were then made to study the influence of 

 small amounts of dissolved electrolytes upon the surface 

 electrification. For this purpose a slightly different form of 

 cell was used, designed to allow the use of larger currents 

 than were possible in the last cell. A straight piece of 

 glass tubing 23 cm. in diameter and about 7 cm. long was 

 fitted with brass stoppers covered on one face with platinum 

 to act as electrodes (fig. 5) . 



Fig. 5. 



i 



Two thin disks of glass mounted just in front of the 

 electrodes prevented small bubbles of gas which might form 

 on these from drifting out into the liquid and disturbing the 

 bubble under observation. For the very dilute solutions 

 that w^ere used this cell was found to be fairly suitable. It 

 was filled and emptied through a small hole in the end. 

 Mathematically (Lamb, Brit. Assoc. Report, 1887, p. 495), in 

 such a closed cylinder the endosmose along the walls would 

 set up a return current along the axis. In a tube of this 

 size, however, it would be negligible, giving an error of the 

 same order as that present in U-tube methods of determining 

 the velocities of colloidal suspensions. 



A considerable amount of work has been done on liquid- 

 solid surfaces and on the effects of dissolved electrolytes on 

 the surface electrification. Hardy (Journ. Phys. Chem. iv. 

 p. 235, 1900) gives as a general rule that, in the case of 

 colloids, the coagulative power of a salt is determined by the 



