On a Modified Water-dropping Influence-machine. 283 



(3) If we now titrate with standard caustic-soda solution 

 equal volumes of the liquid in the collecting- vessels and of the 

 main bulk of the electrolyte, we ought to find that the former 

 is the more acid. 



It may, however, be noticed that, assuming the streaks were 

 not concentrated acid, we should collect by this arrangement, 

 not only the streaks, but also some of the concentrated acid 

 which is formed at the anode. Thus, even if the acid in the 

 collecting-vessels did become stronger, it would scarcely prove 

 that the streaks were the cause of this*. 



From the experiments detailed above, though direct proofs 

 are apparently not available, we may draw the following general 

 conclusions: — That when gas is not evolved at an electrode, 

 streaks are formed there. These are due either to a concen- 

 tration or weakening of the electrolyte, as in the case of 

 solutions of acids and alkalies and some salts (such as zinc 

 sulphate); or, in the case of other salts (such as potassium 

 sulphate), to a chemical change in the electrolyte, yielding at 

 one pole alkali and at the other pole acid, producing solutions 

 of different density to the electrolyte. When gas is not 

 evolved at an electrode these changes are still produced, but 

 their effect in producing streaks is destroyed by the evolution 

 of the gas. 



In our next paper we hope to describe some experiments in 

 which these effects become of great importance in changing 

 the resistance of the electrolyte. 



XXXVI. On a Modified Water -dropping Influence-machine. 

 By Professor Silvanus P. Thompson, D.Scj 



THE ordinary form of water-dropping influence-machine, 

 as devised by Sir William Thomson |, possesses some 

 inconveniences : it requires a double jet of water and special 

 arrangements for high insulation. A simpler form, requiring 

 but one water-jet and mere silk strings (well paraffined) as 

 insulators, has been found by the author to give far less trouble, 

 and to work well for lecture demonstrations. 



* Some observations have been made by this method, but the results 

 have not been very definite. We have designed an automatic commu- 

 tator which will reverse the current at intervals of 15, 30, or 60 seconds. 

 This arrangement will enable us to collect more easily the substance 

 forming the streaks. 



t Communicated by the Physical Society : read January 28, 1888. 



j Proc. Roy. Soc. June 20, 1867 ; and Reprint of Papers on Electro- 

 statics, p. 321. 



