1896.] Mercurons cJihride. 87 



These results closely agree not only among themselves but also 

 with the theoretical quantity. But they appear to be very slightly lovv^ 

 The reason probably appears to be this. Mercurous oxide dissolves to 

 a small extent in water and caustic soda solution. When, however, 

 such a solution is acidified with dilute nitric acid, slight opalescence 

 is sometimes noticed. On subsequent addition of silver nitrate, silver 

 chloride is precipitated, and this cariies down with it traces of mercurous 

 chloride. Silver chloride before it is weighed is heated to incipient 

 fusion, a temperature sufficient to volatilize traces of mercurous chloride. 

 Thus a trace of chlorine is lost. Sufficient stress cannot be given to 

 this explanation, as the difference between actual and theoretical quan- 

 tities can be reduced to a trifling amount by careful manipulation. 



It is evident from the above experiments in conjunction with those 

 mentioned before, that neither gentle warming nor prolonged contact 

 in the cold is sufficient to bring about complete decomposition of the 

 m.ercurous chloride by solution of caustic soda. The coating of mer- 

 curous oxide and the repellent tendency of mercurous chloride to 

 become moistened with water prevent intimate contact with the solu- 

 tion. Complete decomposition is only obtained when the solution is 

 heated to the boiling point and, to be certain, repeating the operation 

 once more with fresh solution of the alkali. 



When the precipitate of mercurous oxide is heated with water or 

 solubility of mercurous solution of caustic soda and then filtered in 

 oxide in water. h^q l,ot state, the filtrate after acidification 



with pure hydrochloric acid is at once turned brownish black when 

 sulphuretted hydrogen is passed through. On heating, the precipitate 

 settles rapidly and this black substance gives all the reactions of 

 mercury. As already mentioned, there was no soluble salt of mercury 

 in the residue. It dissolved with slight separation of mercury in pure 

 dilute nitric acid and this solution gave no precipitate, not even opal- 

 escence with silver nitrate. It is therefore evident that the whole of 

 chlorine must have been eliminated in the form of soluble chloride, and 

 the presence of mercury in the filtrate is due to the solution of mer- 

 curous oxide in water (mercuric oxide and metallic mercury, the pro- 

 ducts of decomposition being practically insoluble, the former 1 in 

 1 50000). In cold water too, mercurous oxide dissolves to a small exten. 



The best method of procedure, therefore, is to decompose the mer- 

 curous chloride by the alkali at the boiling 

 Summary. temperature but to filter when cold and wash 



with cold water. Volumetric estimations of chlorine and silver have 

 been tried but they appear to me to be less trustworthy than these 

 estimations obtained gravimetrically. 



