898 



Foote and Langley — Indirect Method for 



strai<>;ht lino indicates that the oxides liave foniied a solid 

 solution and not a mechanical mixture. The irregularities in 

 the cnrve, which can hardly be due entirely to experimental 

 error, suggest that the oxides are not soluble in each other in 

 all ]>roportions, but that tantalum oxide has taken up about 

 IS per cent of its weight of columbinm oxide, and the latter 



Fig. 1. 



•5p^,9 





ctn 



10 



zo 



30 



40 



SO 



60 



70 



80 



90 



TaO 



over twice its weight of tantalum oxide. The intermediate 

 mixtures would then consist of mechanical mixtures of each 

 oxide saturated with the other. In these mixtures the varia- 

 tions in densities of different mixtures of the same composition 

 are greatest. Points on this part of the curve should lie on a 

 straight line if the interpretation of these results as suggested 

 is coi'rect. 



Analyses of Minerals by Density Method. 



The analysis of a sample of stibiotantalite from the Brush 

 collection was carried out using the indirect method for deter- 

 mining the relative proportion of tantalum and columbium 

 oxides. The density of the sample was 6*80 and the weight 

 four grams. The method used was as follows : Two one gram 

 samples and a two gram sample were each dissolved in about 

 20'='^ of hydrofluoric acid. Some white insoluble material 

 remaining undissolved was filtered off and its weight deducted 

 from that of the sample taken. The resulting weight was 

 taken as that of stibiotantalite used for analysis. The solution 

 containing an excess of hydrofluoric acid was diluted to .300*^° 

 and hydrogen sulphide passed in until all antimony and bis- 

 muth were precipitated as sulphides. The precijiitate was 



