Thornton, Jr. — Separation of Titanium from Iron, etc. 411 



results of two experiments. Since experiment (14) was en- 

 tirely satisfactory, it was deemed unnecessary to extend the 

 series further. 



It only remained to connect the three separations in order to 

 complete the method of analysis. In experiments (15), (16), 

 (17), and (18) known quantities of iron were introduced by 

 adding weighed portions of Kahlbaum's ferrous ammonium 

 sulphate. In experiments (19) and (20), a standard solution of 

 ferric ammonium sulphate was used. In one portion (a) of 

 25 cm3 the iron was determined by titration with potassium per- 

 manganate after reduction of the iron by zinc — the potassium 

 permanganate having been previously standardized against 

 sodium oxalate. In another portion (b) of 25 cm3 the iron was 

 determined by precipitation with ammonium hydroxide and 

 ignition of the ferric hydroxide to ferric oxide. Parallel deter- 

 minations gave the following results : 



Ferric sulphate soln. Ferric oxide 



(a) 25 cm3 0-2267 grm. 



(b) 25 cm3 0-2269 grm. 



The value obtained in (a) was taken as correct. Throughout 

 the fourth series of experiments a standardized solution of 

 phosphoric acid was used. This was prepared by diluting 

 Baker and Adamson's 85 per cent acid. In two portions of 

 25 cm3 each the phosphoric acid was precipitated as ammonium 

 magnesium phosphate. Duplicate determinations gave the fol- 

 lowing results : 



Phosphoric acid soln. Phosphorus pentoxide 



(a) 25 cm3 0-15256 grm. 



(6) 25 cm3 0-1528 grm. 



The mean of these two values was taken as correct. Since the 

 phosphoric acid present in titaniferous rocks is generally small 

 in amount relative to the other constituents, it was thought best 

 to dilute this solution ten fold by volume. The resulting solu- 

 tion, therefore, contained in 25 cm3 0*0153 grm. of phosphorus 

 pentoxide. In Table IY are set forth the results of six experi- 

 ments in which the whole analysis was performed. 



For the benefit of the practical analyst the technique of the 

 entire analysis is here connectedly given. To the solution, 

 which should not be much greater than 100 cm3 in bulk, at least 

 four times the aggregate weight of the four oxides to be held 

 by it in solution of tartaric acid is added. To facilitate the 

 reduction of the iron the solution is made neutral with 

 ammonium hydroxide and acid again with 2 cm3 of sulphuric 

 acid (1:1). Hydrogen sulphide is then introduced until 



