402 



PROCEEDINGS OF THE NATIONAL MUSEUM. 



VOL. 57. 



for an analysis. A very minute amount of pure material gave 45.77 

 per cent of silica, strong qualitative tests for ferric iron, no ferrous 

 iron, and very small amounts of alumina and magnesia. In one or 

 two specimens the interiors of lumps of this golden substance were 

 still green and contained ferrous iron. Later a specimen was found 

 in the collection of William Fitts, of Springfield, which contained 

 more of the yellow mineral than any found at the quarries. Some 

 small lots of the mineral taken from this specimen were analyzed, and 

 the results of the analyses after deducting calcite indicated that it 

 was essentially of the composition of stilpnomelane in which the fer- 

 rous iron is completely oxidized to ferric iron and the hydration is 

 increased by the addition of hygroscopic water. It will thus be seen 

 that in physical characters and chemical composition it is essentially 

 identical with the chalcodite of Shepard.* Shepard included under 

 this head also the associated green material high in ferrous iron 

 which Brush ^ has shown to correspond to stilpnomelane. Brush 

 attempted to include all chalcodite under stilpnomelane, but it now 

 appears that stilpnomelane is the wholly ferrous chlorite, while the 

 varietal name chalcodite should be restricted to the golden ferric 

 alteration product and indeterminate intermediate minerals con- 

 taining varying amounts of iron in both states of oxidation are 

 merely transition products in the change of stilpnomelane to chalco- 

 dite. The samples analyzed gave the following results: 



Analyses of chalcodite, Westjield. 



The material was not of sufficient purity nor in sufficient amount 

 to permit very exact analytical work, and the results give no suffi- 

 ciently reliable basis for speculation as to the exact chemical nature 

 of or the correct formula for chalcodite. For the present it must be 

 regarded as an indefinite oxidation product of stilpnomelane. Under 

 the microscope the material is micaceous and partly transparent. 

 The scales lie upon the flat face and are dark in all positions between 

 crossed nicols. The mean index of refraction is about 1.64, but 



» Shepard, Rep. Arner. Assoc. Adv. Sci., vol. 6, 1851, p. 232. 

 » Brush, Amer. Joum. Sci., vol. 26, 1858, p. 198. 



