﻿104 Mr. J. A. Phillips on the Chemical Composition and 



In order to determine the respective amounts of ferrous and 

 ferric oxides present, separate fusions were made, in a covered 

 platinum crucible, of weighed quantities of pulverized rock mixed 

 with six times their weight of carbonate of sodium, the surface 

 of the mixture being covered by a thick layer of pure carbonate 

 of sodium. 



The contents of the crucible were thus both fused and allowed 

 to cool in an atmosphere of carbonic acid, produced by the de- 

 composition of the alkaline carbonate by silica and silicates, 

 besides being preserved from oxidation by a fused covering of 

 carbonate of sodium. 



The crucible and its contents were now introduced into a 

 beaker containing hot dilute hydrochloric acid which had been 

 previously boiled, and the whole covered by a glass plate. As 

 soon as bubbles of carbonic acid had ceased to be evolved the 

 solution was allowed to cool, and the amount of ferrous oxide 

 estimated by a standardized solution of bichromate of potassium. 



The ferrous oxide thus found was subsequently converted by 

 calculation into Fe 2 O 3 , and deducted from the total amount of 

 iron estimated as ferric oxide. 



The lime was obtained from the nitrate from the precipitate 

 by acetate of sodium, by the addition of ammonia and oxalate of 

 ammonium, and was weighed as carbonate of calcium. To the 

 filtrate from the calcium salt, phosphate of sodium was added for 

 the precipitation of magnesia as pyrophosphate of magnesium. 

 Titanic acid was estimated by the process recommended by 

 Riley*, which was conducted as follows. 



The silica (a, h, and c) from both analyses was attacked in a 

 platinum dish by hydrofluoric acid, and sulphuric acid added, 

 evaporated to dryness, and ignited. In this way a residue was 

 obtained, which was fused with bisulphate of potassium and dis- 

 solved in cold water; the liquid after complete solution was 

 boiled ; on boiling briskly for some time a precipitate was formed, 

 which was separated by filtration, washed, dried, and ignited. 

 This precipitate gave the characteristic reactions of titanic acid, 

 and was deducted from the total weight of silica previously de- 

 termined. The quantity of titanic acid in these rocks is so ex- 

 ceedingly small, that it was found no appreciable quantity could 

 be obtained byre-solution of the iron, separated from alumina by 

 caustic potassa, boiling with sulphite of sodium, &c, as recom- 

 mended by Riley. Care was therefore taken to evaporate twice 

 to dryness from silica ; and any traces of titanic acid that might 

 be retained by the iron (after separation of silica c) were, ex- 

 cepting in the analyses of slate from Delabole, neglected. 



It having been found that the whole of the phosphoric acid 

 * Quarterly Journal of the Chemical Society, vol. xii. p. 13. 



