ON PYROXYLIC SPIRIT, &c. 129 



tempt to answer it ; but I confess it appears to me to admit of a reply. The in- 

 fluence which water has on a variety of chemical phenomena has been shewn 

 by M. Pelouze, and that of quantity and cohesive force was long ago pointed out 

 by Berthollet ; and in regard to the phenomena of solution, the operation of 

 quantity is generally acknowledged. By the aid of such principles, the appear- 

 ances under consideration seem to admit of explanation. Since every substance 

 is dissolved, in consequence of a chemical affinity between it and the solvent, it 

 follows, that if a chloride is dissolved as a muriate, the affinity of water for the 

 muriate has a considerable share in determining the union of the metal with oxy- 

 gen, and of the chlorine with hydrogen. Now, when the water of such a solution 

 has evaporated away to such an extent as to leave a completely saturated solu' 

 tion, the affinity of the oxygen and hydrogen of the muriate for one another is re- 

 sisted by the attraction of a much smaller quantity of water for the muriate than 

 formerly, and is now aided by the incipient cohesive attraction of the salt in the 

 act of crystallization, joined to the powerful affinity of chlorine and sodium for 

 one another. The affinity of the elements of water, therefore, under the circum- 

 stances in which they now come to he placed, and not its tension before its exist- 

 ence, I should humbly think is the cause of the union of these elements ; and this 

 affinity may very well come into play under the circumstances mentioned^ before 

 the vaporizing tendency of the remaining actual water takes full effect ; and it 

 farther seems probable, that the affinity of the remaining muriate for water will 

 aid this union of the oxygen and hydrogen of the salt in the act of crystallization, 

 the water formed by the union of the oxygen and hydrogen of the salt probably 

 uniting with the dissolved muriate before its dissipation by evaporation. 



In so far, therefore, as I am able to judge, it does not appear that any of these 

 ai'guments foreclose the inquiry ; and we may still be at hberty to bring forward 

 illustrations on the other side. 



Let us assume for a moment, that chlorides and iodides are dissolved as such 

 in absolute alcohol, and as muriates and hydriodates in water. Let us next sup- 

 pose these alcoholic and aqueous solutions exposed to voltaic agency, under cir- 

 cumstances in which no secondary action can take place at the poles from emhed oxy- 

 gen and hydrogen. What ought to happen in the two cases of alcohohc solution 

 and of aqueous solution ? Surely this. In the alcoholic solution, since it has been 

 shewn, I hope successfully, that water only is directly decomposed, then neither 

 constituent of the chloride or iodide, nor any acid, ought to be produced in that 

 solution ; whilst in the aqueous solution, the salt composed, ex hypothesi, of acid 

 and alkali, ought to be resolved into its elements, conformably to the general law 

 of the electric decomposition of ordinary salts, and acid and alkali, without chlo- 

 rine or iodine, should appear in the solution, each on its proper side travelling to- 

 wards the poles, which, by the supposition, are placed beyond the bounds of that 



VOL. XIV. PART I. B 



