150 Graton and Schaller — Purpurite, a new Mineral. 



(1/5 gram) and the many operations necessary, the agreement 

 is as close as could be expected. The average of the two 

 results is 29*20 per cent. 



A direct determination of the total manganese, weighed as 

 anhydrous sulphate, gave as the amount of Mn 2 3 in the min- 

 eral, 29*35 per cent, which agrees almost exactly with the 

 average of the two indirect determinations. 



The remaining constituents were determined as follows : A 

 portion of the mineral was dissolved in hydrochloric acid and 

 a known weight of iron added (as ferric chloride). A basic 

 acetate separation w r as then made, boiling the solution for fif- 

 teen minutes, which according to Bunsen will precipitate all 

 the phosphoric acid with the iron and will not precipitate any 

 manganese. The precipitate was dissolved in hydrochloric acicl, 

 and reprecipitated by ammonia, after the addition of some 

 ammonium chloride. The two filtrates were united, man- 

 ganese precipitated by hydrogen sulphide and finally weighed 

 as anhydrous manganese sulphate. Calcium was then thrown 

 out, dissolved and reprecipitated and magnesia found to be 

 absent. The iron-phosphate precipitate was dissolved in 

 hydrochloric acid and divided into two portions. In the one, 

 the iron and phosphoric acid were precipitated by ammonia 

 and weighed. This was then fused up with sodium bisulphate 

 and tested for manganese with silver nitrate and ammonium 

 persulphate. None was present. In the second portion, the 

 iron was reduced by hydrogen sulphide and titrated with per- 

 manganate. Phosphoric acid was determined in the usual way 

 and a second value obtained by the difference between the iron 

 and the iron plus phosphoric acid. The alkalies were deter- 

 mined by the Lawrence Smith method. The final solution of 

 chlorides gave a strong spectroscopic test for lithium. The 

 water below 105° was determined directly, using a toluene 

 bath. The total water was determined directly by heating in 

 a glass tube according to Penfielcl. The water is all given off 

 at a low temperature, that at 105° being given off very readily, 

 and at one time. Further heating at 105° failed to remove 

 any more. The values obtained are as follows : 



Av. Ratio. 



Fe 2 3 15-89, 15-89 1-03) 



Mn 2 3 _-. 29-35,30-47, 27*93 29"25 1-93 ( 



P 2 5 . 47-64,46-96 47-30 3'47 



H 2 5-26 5-26 3'04 



CaO 1-48 1-48 -27) 



Na 2 -84 -84 -14 j 



Li 2 tr. tr. 



Insol '52 -52 



100-54 



