Graton and Schaller — Purpurite, a new Mineral. 149 



tion is probably rather high. It is hoped that sufficient of this 

 material for analysis will soon be obtained. 



The occurrence of purpurite in material collected from 

 California by Mr. Schaller throws additional light on the 

 origin and association of this mineral. It occurs with a black 

 material which appears to be identical with that described 

 above, and both are undoubtedly decomposition products of 

 the accompanying triphylite, the iron-rich member of the 

 lithia-manganous-ferrous phosphate series, of which lithiophi- 

 lite is the manganese-rich end. 



The small number and rarity of minerals containing man- 

 ganic oxide, Mn 2 3 , may be due to the relative instability of 

 that base in comparison with manganese dioxide. 



Chemical Composition ; Waldemak T. Schallee. 



About a gram of pure material was separated by Mr. Graton. 

 This was divided into several portions, using about a fifth of a 

 gram for each determination. The most interesting part of 

 the analysis was to determine the state of oxidation of the 

 manganese. When the mineral is treated with hydrochloric 

 acid, chlorine is readily given off. The manganese present 

 can therefore not be in the manganous state, and the absence 

 of ferrous iron and the presence of ferric iron suggested that 

 the manganese was present as a manganic salt. Such was 

 found to be the case. 



A fifth of a gram was dissolved in sulphuric acid with a known 

 amount of ferrous ammonium sulphate. All precautions were 

 observed to avoid the presence of air, the entire operation being 

 conducted in an atmosphere of carbon dioxide. The water 

 used had been boiled and cooled out of contact with air. Just 

 before the iron sulphate was introduced into the flask contain- 

 ing the mineral, an equal quantity was removed from the stock 

 solution and titrated with permanganate. Thus, the amount 

 of ferrous iron introduced into the flask with the mineral was 

 known. After the mineral had been decomposed by the sul- 

 phuric acid, the flask was cooled and the solution titrated, the 

 amount of iron sulphate oxidized by the liberation of oxygen 

 from the mineral being determined in this way. From these 

 data the amount of Mn 2 3 was calculated and found to be 

 30*47 per cent. 



A second sample was decomposed by hydrochloric acid and 

 the chlorine evolved passed into a solution of potassium iodide. 

 The liberated iodine was then titrated with sodium thiosul- 

 phate, the latter being standardized with pure copper. Cal- 

 culating from the results obtained, the amount of Mn 2 3 was 

 found to be 27*93 per cent. Though these results vary some- 

 what, yet, considering the small amount of material used 



