Influence of Voltaic Currents on the Diffusion of Liquids. 81 



phosphoric acid, by being too viscous, as well as badly conducting, 

 prevented the effect. 



Both the absolute and relative degrees of adhesion of the two liquids 

 to the tubes, especially in very narrow ones, must also affect the results. 

 In a separate paper on the " Phenomena of the Capillary Electroscope," 

 " Proc. Roy. Soc," No. 209, 1881, I have shown that adhesion per- 

 forms an important part ; and that mercury flows more rapidly from a 

 capillary glass tube into dilute sulphuric acid when an electric current 

 is passing in the same direction than when it is not passing ; and it 

 would be desirable, therefore, to examine the influence of adhesion 

 upon the rates of flow of electrolytes (as well as of mercury) through 

 tubes of glass (and other material) whilst under the influence of an 

 electric current. 



That the phenomena of lines and movement of mass of liquid in 

 these experiments are not wholly capillary is proved by the circum- 

 stance that they have been produced in tubes of as large a diameter 

 as 15 millims. Capillary action, however, affects the results (see 

 p. 78). 



Assuming that the explanation I have offered of the phenomena 

 (see p. 67, et seq.) is correct, ordinary diffusion is affected during 

 the passage of the current, the whole of the molecules being subjected 

 to electric influence, which probably aids diffusion at the one meniscus 

 and neutralizes it at the other. A result obtained in the experiments 

 on electric osmose, viz., that the liquid flow is more rapid when the 

 direction of electric osmose coincides with that of ordinary osmose, 

 than when it is opposite to it, supports the conclusion that ordinary 

 diffusion (although suspended) influences the results. In Experiment 

 No. 15 the dilute nitric acid diffused into the cupric bromide ; this 

 diffusion was by the influence of the current decreased at the negative 

 meniscus and increased at the positive one ; an opposite effect to that 

 exerted by the current upon water. The interference of ordinary 

 diffusion might probably be largely obviated by using liquids of equal 

 specific gravity. 



The less rapid osmose which occurs where an electric current passes 

 from a strong (through a diaphragm) to a weak solution, than where it 

 passes in the reverse direction, is not likely to be due to difference of 

 chemical action, because at each junction of the liquids in the cells 

 divided by porous partitions (see " Proc. Roy. Soc," vol. 31, p. 253), 

 there was opposed equivalent quantities of acid and bases set free by 

 electrolysis, and therefore equal chemical action. Much more investi- 

 gation, however, remains yet to be made in this part of the subject. 

 Although the experimental evidence of an influence of the current upon 

 ordinary diffusion without a porous diaphragm is not extensive, it is 

 sufficient in Experiments Nos. 40, 41, 42, and 43 to prove that diffusion 

 of liquids is greatly affected by the passage of a current, and that it 



VOL. XXXII. g 



