10 Dr. Kane on ike Compounds of Ammonia. 



also, it appeared from the evidently inconstant nature of the subnitrates obtained 

 by water, that the stages of its production required to be closely studied. In 

 addition I had observed that nitrous turpeth, when heated, always yielded some 

 liquid nitric acid ; this fact should introduce water as one of its constituents, 

 which the results obtained by Braancamp and Grouvelle necessarily exclude. 



A quantity of crystallized nitrate of mercury was treated by water, and the 

 undissolved portion washed by warm water until the washings no longer reacted 

 acid. It then appeared as a fine yellow powder, very heavy, not acted on by cold 

 water, but converted into a brownish red powder by boiling water, which dis- 

 solved out the soluble nitrate of mercury, not affecting blue cabbage paper. 

 When this powder is heated, it gives much red fumes and a quantity of liquid 

 nitric acid, and there remains red oxide of mercury, which by a stronger heat is 

 decomposed. As, by avoiding the use of boiling water, this powder was obtained 

 apparently similar in appearance and properties at different times, it was selected 

 for analysis. 



A. 5.458 grammes of this powder were dissolved in muriatic acid, and 

 treated by proto-chloride of tin. There were obtained 4.170 grammes of metallic 

 mercury, giving 76.40 ug per cent. 



B. 5.513 grammes of a portion prepared at a different time were dissolved 

 in muriatic acid diluted with a good deal of water, and precipitated by sulphu- 

 retted hydrogen ; there were obtained 



Filter and sulphuret = 5.935 1 f. ^^o 



Filter = 0.932 / ^ 



giving mercury =. 78.31 per cent. 



C. A portion of the yellow powder having been treated by boiling water, 

 and having assumed a brownish red colour, was dissolved in muriatic acid, and 

 precipitated by sulphuretted hydrogen. Thus analyzed, 4.975 of this powder 

 gave 4.919 sulphuret of mercury, corresponding to 85.33 mercury per cent. 



D. A quantity was boiled for a long time, until it had been converted into a 

 brick red powder, which was analyzed by solution in muriatic acid and the sepa- 

 ration of the mercury by proto-chloride of tin ; from 7.746 grammes were ob- 

 tained 6.673 mercury, or 86.17 per cent. 



No matter how far the boiling might be carried, I could not reduce the 

 powder to thq state of pure red oxide. The residual powder dried always gave 



