184 Hartung and Rivett : 



would undergo chemical action with sodium chloride, the presence 

 of which in a dried-up lake near the sea would be probable. 

 Ammonium chloride would result, and at the temperature of the com- 

 bustion would be volatilised. A number of analyses have been 

 carried out in order to ascertain definitely the nature of the materials 

 taking part in, or produced during, the reactions which occur. The 

 following paragraphs summarise the results. 



1. Specimens of the crystalline sublimate from one of the vents 

 contained 99.4 per cent, pure ammonium chloride, calculated from 

 the weight of silver chloride obtained by double decomposition with 

 silver nitrate. The balance was probably moisture, or a trace of 

 contaminating ash. It contained no sulphate. A natural product 

 from Vesuvius, analysed by Klaproth, contained 99.5 per cent, 

 ammonium, chloride, and 0.5 per cent, ammonium sulphate. 



2. Specimens of red ash gave the following proportions of the 

 main constituents : — 



99.4 99.8 



Alkalis were not determined quantitatively. Sodium was present 

 in small amount, and there was a minute trace of potassium. Quali- 

 tative tests showed that manganese, zinc, cobalt, nickel, barium, 

 strontium, chromium, carbonate and chloride were absent. The 

 complete absence of chloride is to be noticed. It is unlikely that 

 sodium chloride would be volatilised during combustion, and unless 

 it were washed out of the layer by heavy rain before combustion 

 began, it is probable that there was more than sufficient ammoniacal 

 matter produced to convert the whole of the chloride to the am- 

 monium salt. This possibility is supported by the fact, to be 

 mentioned later, that the yield of ammonium chloride may be 

 increased by addition of common salt to the combustible matter. 

 A small amount of soluble matter could be extracted by water from 

 liif ash. [t contained calcium, magnesium and sulphate. On treat- 

 ing the ash with boiling hydrochloric acid, chlorine was evolved. As 

 higher oxides of manganese were absent, one is inclined to suppose 



