1876.] and Electrolysis in Chemical Compounds, 327 



If the same current was conducted through pure water, immediately 

 a copious quantity of gas was evolved, but nothing of a vibratory motion 

 in the liquid was visible. This seems only to occur in the case of highly 

 insulating liquids, and illustrates the conduction of electricity in the way 

 which was called by Taraday " the carrpng discharge ; " this acts me- 

 chanically; and it appears that the greater part of electricity is then trans- 

 ported by the molecules without their splitting up, as in the case of 

 electrolysis. 



The results of the experiments carried on in this investigation, with 

 regard to electrolysis, may thus be stated : — 



1. A general connexion between definite chemical characters and lia- 

 bility to electrolysis seems not to exist. The character that a compound 

 contains hydrogen, which is easily replaced by a metal or radicals, or 

 that it contains a metal for which another may be easily substituted, is 

 not always accompanied with the conductive power that is required for 

 electrolysis. With liquid ammonia this is the case, but with zinc-ethyle 

 (which in contact with air bursts into flames, and is destroyed from its 

 Zn changing into ZnO), benzine (which is easily transformed in different 

 compounds by substitution), and tetrachloride of tin an enormous resist- 

 ance is offered to the most powerful galvanic current ever used. 



2. Sometimes a very bad conductive power is accompanied by a great 

 difficulty in getting the hydrogen replaced in the compound. A re- 

 markable example of this fact is offered by liquid hydrochloric acid ; 

 this compound was kept for months, even years, in contact with strips of 

 zinc, and up to the present time very little action is perceptible. 

 Gore* has communicated several experiments of this kind with similar 

 results. 



3. It appears that it is not the nature of the actual constituents in the 

 compound which renders it proper to conduct electricity ; but that this 

 is more dependent on the inner arrangement of the molecules. 



4. Although in the case of very bad conductors, as liquid carbonic acid 

 and liquid hydrochloric acid, no test for electrolysis can be applied, 

 yet it must not be concluded that they may not be decomposed by elec- 

 trical agency. By using spirals of zinc and platinumf twisted together 

 for a considerable space of time, they were actually split up ; tins fact 

 shall be further elucidated in the paper mentioned in § 1. 



The Hague, September 1876. 

 * Phil. Mag. [IV.] vol. xxix. p. 543 (1865). 



t Gladstone and Tribe used this combination to analyze water and some organic 

 compounds (Journal of the Ohem. Society, 1872, p. 461). 



