306 EXPERIMENTS AND OBSERVATIONS 



acid gas, this may not prevent the formation of a super-muriate, 

 more especially as the excess is in the metallic form, and ex- 

 erts no direct action, therefore, on the real acid. 



To ascertain if a super-muriate were formed in these cases, 

 the product obtained from the action of the muriatic acid on 

 the metal was raised to a heat as high as could be applied with- 

 out volatilization, so that no loosely adhering acid might re- 

 main, and the air in the retort was repeatedly drawn out by a 

 caoutchouc bottle. The solution from the residue both of iron 

 and zinc was very sensibly acid. Some fallacy, however, at- 

 tends this, from the circumstance, that the liquid state is ne- 

 cessary to admit of the indications of acidity, and in adding wa- 

 ter to produce this, a change occurs in the state of combina- 

 tion, in a number of the metallic muriates ; a super-muriate be- 

 ing formed, which remains in solution, and a sub-muriate being 

 precipitated, so that the acidity of the entire compound cannot 

 justly be inferred from that of the solution. I found, accord- 

 ingly, that on adding water to the product from the action of 

 the acid gas on zinc, this change occurs ; a little of a white pre- 

 cipitate being thrown down, while the liquor remained acid. 

 But the fallacy can be obviated, by adding only as much water 

 as produces fluidity, without subverting the combination. Por- 

 tions, therefore, of the residue were exposed to a humid atmo- 

 sphere, until by deliquescence, liquors were formed transpa- 

 rent, without any precipitation ; and these were strongly acid, 

 reddening litmus paper when it was perfectly dry and warm. 

 I farther found, that the product of the solution of zinc in li- 

 quid muriatic acid, when digested with an excess of metal, and 

 evaporated to dryness, afforded by deliquescence a liquor sen- 

 sibly acid. And in both cases, even when the solid product 

 was retained liquid by heat, acidity was indicated by litmus pa- 

 per. Lastly, What is still less liable to objection, the residue 



in 



