ON PYROXYLIC SPIRIT, &c. 127 



the latter, apparently owing to an action of the nature of endosmose; in which 

 way a little of the salt itself may also be carried over, so as to augment the secon- 

 dary action. 



Alcoholic solutions of chlorides are less well adapted for such experiments 

 than those of iodides, because if chlorine were evolved in an alcoholic liquid, we 

 know that it would immediately react on the alcohol, giving rise to muriatic acid 

 and other products. Still, if such a solution were connected in the usual way 

 with water, some portion of the chlorine, if directly produced, might perhaps 

 escape this reaction. It is at least proper to state the result actually observed, 

 when it will appear that nothing contrary to the idea of a secondary action was 

 noticed. 



Absolute alcohol containing i^gth of recently ignited chloride of calcium was 

 placed in a tube A of one and a half dram capacity, and connected by asbestus 

 with water in a similar tube B, as in Fig. 2, the former liquid being made nega- 

 tive and the latter positive, by a power of 72 pairs of 4-inch plates. In four 

 minutes acid appeared at the positive pole, and an alkaline reaction at the nega- 

 tive, with effervescence from both poles, but no smeU of chlorine was perceived. 

 After half an hour's action, there was still no smeU of chlorine in either tube, nor 

 any bleaching action, whilst the positive liquid was acid, and the negative showed 

 an alkaline reaction, and the negative foil was coated with lime. On reversal, no 

 chlorine was disengaged in the positive liquid, because it immediately reacted on 

 the alcohol, which in consequence soon became strongly acid. In short, this ex- 

 periment, if it affords no positive evidence in favour of a secondary action, is at 

 least perfectly explicable on that idea. 



Before proceeding to draw these general conclusions as to the nature of the 

 voltaic decomposition of solutions in different solvents, to which I alluded in the 

 outset of this paper, I think it better to describe those experiments which appear 

 to illustrate the states in which haloid salts exist in solution in alcohol and 

 in water, because additional evidence will be, in the course of them, afforded 

 of the secondary origin of the electro-negative constituent of such salts in the 

 electric decomposition of their solutions, and because we shaU be better able to 

 draw the conclusions referred to when the nature of such solutions has been 

 examined. 



IV. — Voltaic Experiments illustrative of the state in which Haloid Salts are dis- 

 solved by water. 



The question, whether chlorides and other analogous salts are dissolved as 

 such by water, or decompose it, and exist in solution as muriates, &;c. remained 

 after the old theory of the nature of chlorine had been abandoned, and the simple 

 nature of that substance had been universally acknowledged. At the present 



