136 MR CONNELL ON THE ACTION OF VOLTAIC ELECTRICITY. 



comprehended within the bounds of the proposed law, for it rarely happens that 

 the inductive process actually embraces every particular, its imperfection in point 

 of logic being usually supplied by the necessary connections of the individual 

 cases. All I can say is, that I am at present acquainted with no exception to the 

 proposed law, and that all the experiments which I have yet made, go to support 

 it ; but stUl as there are some cases comprehended in it, Avhich may and ought to 

 be experimentally investigated, I shall not yet take upon me to give it as esta- 

 bhshed in its utmost generality, but shall probably in a futm-e communication 

 state the farther results obtained. 



This rule is of course entirely confined to compounds of elementary bodies. 

 Every one knows that an ordinary salt dissolved in water, is resolved into its 

 constituent acid and alkali under voltaic agency. The same observation I have 

 found to apply to alcoholic solutions of such salts. With respect to their ethe- 

 rial solutions, it would seem that it does not hold, in so far as reMance can be 

 placed on a single experiment with a moderate voltaic power. A solution of ni- 

 trate of uraniimi in rectified ether was submitted in a close tube to the action of 

 fifty pairs of 2-inch plates, without any appearance of the constituents of the 

 salt at their respective poles, or action on the galvanometer formerly described. 



ERRATA in former Memoir in Vol. XIII. ol Edinburgh Transactions. 



Page 316 (p. 2 of separate Memoir), line 20, for liquid read soUition 



— 334 (p. 20 — — — ), — 8, _ ellier — alcohol 



— 337 (p. 23 — — — ), note, line 4, for effects read quantities 



— 346 (p. 32 — — ), lines 8 and 9 for positive read negative, and for negative read positive 



