ON PYROXYLIC SPIRIT, &c. 119 



positive extremity of the asbestus, with efifervescence from both poles ; and in 

 about eight minutes a slight odour of chlorine was observed, and also acid reac- 

 tion at the positive side of the liquid in A. In a quarter of an hour test-paper 

 was not bleached when dipped into either tube. After upwards of twenty mi- 

 nutes, no smeU of chlorine could be distinguished in A, and when the action was 

 then suspended the odour in B was stni only slight. The battery was now re- 

 versed as before described. An instant smell of chlorine arose, with effervescence 

 from the positive pole, and test-paper was bleached there within one minute. 

 Effervescence also from the negative pole. 



A moderately strong solution of iodide of potassium was now substituted for 

 chloride of potassium, all other circumstances, including the voltaic power, being 

 the same as at the commencement of the preceding experiment. In five minutes 

 there was acid reaction at the positive side of the Uquid in A ; and about the same 

 time browning was observed to be just beginning in the positive licpiid near the 

 termination of the asbestus, and extending to the positive foil. This bro^vning 

 went on increasing in the positive liquid, with acid reaction on the neighbom'ing 

 asbestus and positive side of the negative liquid. The colour of the negative li- 

 quid was not changed. In this particular experiment the battery was not rever- 

 sed ; but in numerous others with the smaller powers, it was always found that 

 the reversal caused immediate production of iodine at the positive pole Avithout 

 effervescence, whilst gas arose from the negative. 



These results are quite conformable to those with the hydracids. Chlorine 

 or iodine appears in virtue of acid passing to the positive side and suffering a se- 

 condary action. The degree of observed secondary action is proportional to the 

 facihty with which the corresponding acid is decomposed by nascent oxygen, and 

 not to the absolute quantity of acid which appears at the positive side, the se- 

 condaiy action being strongest in the case of iodide of potassium, whilst the quan- 

 tity of acid on the positive side is much smaller than in the case of chloride of po- 

 tassium. I shall afterwards describe an experiment similar to that with hydrio- 

 dic acid and two other vessels of water, in confirmation of these views. I shall 

 then also state the grounds which have led me to infer from the appearances un- 

 der galvanic agencj', that haloid salts do not exist in solution as such, but as hy- 

 dracid salts ; in other words, that when dissolved they decompose water. 



It was shewn a few years ago by M. de la Rive, that when a mixed solution 

 of bromide of iodine and stai'ch in water was acted on voltaicaUy, iodine appeared 

 at the positive pole, and formed the usual blue combination with the dissolved 

 starch.* By an experiment conducted on similar principles with those just de- 

 scribed, I have been led to conclude, that the action is here also a secondary one. 

 The mixed solution placed in a tube was connected with the positive side of fifty 



* Annales de Chim. et de Phys. xxviii. p. 160. 



