1880.] 



Experiments on Electric Osmose. 



253 



liquid were produced without electrolysis, and electrolysis would 

 reasonably account for several of the phenomena, it was concluded 

 that electrolysis was not merely a concomitant circumstance, but also 

 acted as a cause. The converse phenomena of the subject of this 

 paper have been made the subject of a separate research, entitled 

 " The Production of Electric Currents by Diffusion and Osmose of 

 Liquids." ~No effect of magnetism upon the lines or movements of 

 the liquids was sought for. 



Chemical action was found to be only a coincident and not a funda- 

 mental part of the phenomena ; it constantly took place by electro- 

 lysis and by contact of liberated ions, and occasionally by contact of 

 the original liquids. The lines and strata of liquid at the meniscuses 

 were produced with almost every possible chemical combination of 

 electrolytes ; and references are given to experiments which prove 

 this ; they were also the more difficult to produce the less the 

 differences of chemical composition of the two liquids. 



VIII. " Experiments on Electric Osmose." By G. Gore, LL.D., 

 F.R.S. Received December 7, 1880. 



The following experiments were made for the purpose of elucidating 

 a question in a research on " The Influence of Voltaic Currents on the 

 Diffusion of Liquids" (ante, p. 250), and are published separately, in 

 order to facilitate reference to them by other investigators. 



As, also, the discovery of an exceptional instance is often of con- 

 siderable importance, an additional number and variety of solutions 

 were purposely examined with that further object until such a one was 

 found. I anticipated that the proportion of exceptional instances 

 would be " about one or two per cent." 



The experiments were similar to that made by Porrett (" Annals of 

 Philosophy," vol. 8, p. 74). The vessel employed was about 7'0 

 centims. high and 3'0 centims. diameter, of thick glass, divided vertically 

 into two equal parts, its edges being covered with vulcanised india- 

 rubber, with a diaphragm of biscuit ware about 1'5 millim. thick 

 placed between them ; the whole being held together between two 

 upright bars of wood by means of a clamp- screw, so that the dia- 

 phragm might be readily removed, and replaced by a clean one. 

 The electrodes were of sheet platinum, about 18 millims. wide ; and 

 the electric current (except in the cases otherwise mentioned) was 

 from 12 Grove's cells of one pint capacity, and in single series. The 

 liquid was usually about 2 centims. high in the vessel, and at the same 

 level on each side of the partition previous to passing the current in 

 each experiment. The diaphragm was frequently changed. 



The following table exhibits the compositions of the liquids tried, 



VOL. XXXI. '£ 



