189 



Concerning the method, in wliich the analysis was made, 

 Christensen states as follows : 



«The mineral was fused with sodium carbonate and 

 evaporated with nitric acid. The mixture of silica, niobic acid 

 and titanium dioxide was ignited and weighed. The silica 

 was expelled with hydrofluoric and sulphuric acid. The silica 

 was then deduced from the difference in weight.» 



"In the mixture of titanium dioxide and niobic acid the 

 titanium was found by dissolving in hydrofluoric and sulphuric 

 acid, by volatilizing the hydrofluoric acid and treating the rest 

 cautiously with cold water, by which process all went into 

 solution, in this the titanium was determined colorimetrieally 

 by adding hydrogen peroxide and comparing the yellow color 

 with a solution in sulphuric acid of pure titanium dioxide 

 produced in the same way.» 



«A small quantity of niobic acid was found in the 



original solution containing nitric acid and could as regards 



the greater part be precipitated by boiling; the remainder 



was found together with the alcalies.» 



Of the 11-01 per cent H.^0 1-75 per cent disappears 



already under 100° and accordingly is probably present as 



hygroscopical water; as the approximate formula is found for 



the epistolite : \%SiO^-\ TiO^ ■ 5 M, 0-^ • [Ca, Mg, Fe, Мп) О 



■ lONa^O-2llL,0-iNaF. 



Since the mineral is possibly somewhat altered and since it 

 is not known, how much H^O belongs to the molecule itself, it 

 will hardly be possible either to fix this formula more definitely 

 or to classify the composition under a certain group of silicates. 

 On account of the small quantity of basic component parts, 

 which it contains, the mineral must be classified among the 

 very acid silicates. 



The epistolite has neither in its chemical nor its physical 

 properties any similarity to any other mineral, it has a separate 

 position since it contains the most niobium of all silicates without 



