430 



PROCEEDINGS OF THE NATIONAL MUSEUM, 



VOL. 41. 



The results obtained as to loss of water at different temperatures 

 are as follows: 



Loss of water on heating. 



*The color called light gray is very difficult to define. The mineral powder had lost its original green 

 color but did not show any decided lavender shade. Perhaps the combination of some lavender and some 

 green gave the resultant effect here called gray. 



The vanadium and clu-omium were determined on a 4-gram 

 portion of the mineral, the method used being that described by 

 Hillebrand.* After the cliromium had been determined colorimet- 

 rically, the correction necessary to apply to the vanadium deter- 

 mination was found by titrating a hot sulphuric acid solution of 

 sodium chromate containing the same amount of chromium oxide 

 (CrjOg), as was present in the sample of variscite analyzed. Qualita- 

 tive tests with ferricyanide showed that there was no ferrous iron in 

 the solution of the mineral which had been heated to 160° C. There- 

 fore, it was possible to determine the state of oxidation of the vana- 

 dium. For this purpose a gram of the mineral was heated to 160° C, 

 which completely dehydrated it and changed its green color to the 

 intense lavender. The powdered mineral was then dissolved in 

 sulphuric acid and titrated with permanganate. The amount con- 

 sumed was equivalent to 0.45 per cent vanadium oxide (V^Og) and, 

 corrected for the chromium present, gave a value of 0.41 per cent 

 vanadium oxide (VgOg). This is higher than the value obtained on 

 the 4-gram sample (0.32 per cent) but the lower result is taken as 

 being the most accurate. It may be here stated that a cold solution 

 of bichromate in sulphuric acid %vill consume a small but appreciable 

 amount of permanganate which has to be determined and allowed 

 for in such instances as the present one. The ferric iron was deter- 

 mined colorimetrically using potassium sulphocyanide. 



An adequate explanation of the remarkable color change and 



accompanying changes in physical and optical properties has not 



been found. 



EXPLANATION OF PLATE 33. 



Upper figure.— Nodule of variscite from Lucin, Utah. Shows the usual compact 

 phase on the left and the crystalline aggregate on the right. Natural size. 



Lower figure. — Crystallized variscite from Lucin, Utah. Shows a cavity in the 

 crystalline aggregate with the tabular crystals. Enlarged 6 diameters. 



Described on page 416. 



» Hillebrand, W. F. The analysis of silicate and carbonate rocks. Bull. 422, U.S. Geol. Survey, 1910. 



