142 THEORETICAL OBSERVATIONS. 



present to substitute for them : we shall therefore prefer leaving 

 them to the elucidation of further investigations to adding one more 

 to the already too numerous list of hasty generalizations." 1 



In this paper, the authors state that they found certain positive 

 elements transferred in small proportions: thus, potassium from 

 sulphate of potash, in -J of an equivalent ; barium, from nitrate of 

 barytes, j? equivalent ; and magnesium, from sulphate of magnesia, 

 i^ equivalent. 



In all cases where two liquids are separated by a porous dia- 

 phragm, there is a mutual transfer of the liquids in distinct ratios, 

 according to time, either by what is called endosmosis, or by a diffu- 

 sion ; and the rate of transfer is materially affected by a galvanic 

 current passing through them. From observations and operations 

 made on a large scale, and from experiments on various kinds of 

 solutions, we are inclined to think that the fractional transfers of 

 Professors Daniell and Miller are the results of endosmosis or diffu- 

 sion rather than of electrolytic transfer. According to recent expe- 

 riments by Professor Graham, diffusion takes place in definite 

 proportions. We are therefore of opinion that no transfer of any 

 base or positive element takes place by electrolysis. 



Proposed Theory. Having carefully considered the various phe- 

 nomena attending electrolysis, in the decomposition of metallic 

 salts, we think that the electricity is conducted through the solu- 

 tion by the base, or positive element, in the electrolyte, which it 

 does as if it was a solid chain of particles or wire. We have 

 already said, that if to a solution of sulphate of copper we put a 

 piece of iron, the acid in union with the copper will leave it and 



combine with the iron. If a piece of copper be put into the same 

 solution, no change will take place ; but if we by any means give 



1 Philosophical Transactions, Part I. for 1844. 



