554 



PROCEEDINGS OF THE NATIONAL MUSEUM. 



VOL. 53. 



ANALYSES. 



Preparation of samples. — Twelve thoroughly clean and fresh repre- 

 sentative samples, eight of rose quartz and four of amethyst, from 

 different localities were selected for analysis. Each sample was 

 crushed without grinding in a steel mortar and passed through 

 silk gauze having 13 meshes to the millimeter. The powder was 

 placed in a porcelain dish and dilute hydrochloric acid added and 

 allowed to stand for about 10 minutes, when it was diluted with 

 water, allowed to settle, and the liquid decanted off. This treat- 

 ment with dilute hydrochloric acid and washing with water was 

 repeated seven or eight times, after which the liquid gave no indica- 

 tion of the presence of iron. The samples were then dried on tho 

 water bath and used for anal3^sis. 



Methods of analysis. — From 4 to 5 grams of the sample, prepared 

 as outlined above, were accurately weighed into a platinum crucible 

 and treated with a few drops of sulphuric acid and hydrofluoric acid 

 and heated until all sihca was volatilized. The residue was fused 

 with sodium carbonate and the melt dissolved in nitric acid. Tho 

 resulting solution was examined for iron, manganese, titanium, and 

 cobalt, according to the methods outhned by Hiliebrand.* 



Iron was determined by titrating with potassium permanganate 

 solution, 1 cc. of which was equivalent to 0.00067 gram of ferric 

 oxide. Manganese and titanium were determined colorimetrically. 

 The precipitate of cobalt sulphide was so small that it was not 

 feasible to handle it gravimetrically, so that the filter containing it 

 was ignited and the total residue taken up in a borax bead. The 

 color of the beads varied from a faint blue to a clear and distinct 

 blue, but in no case was the bead highly colored. 



The results obtained on analysis follow in tabular form : 



Partial analyses of amethyst and rose quartz. 

 [R. E. Beard, analyst.] 



No. 



Variety. 



Constituents determined. 



TiO, 



MnO. I F02O3. 



CoO.« 



0. 00199 

 .00166 

 . 01)145 

 . 00521 

 .00577 

 . 00139 

 .00269 

 . 00288 

 . 00288 

 . 004-15 

 . 00289 

 . 00293 

 .06900 



0. 00036 

 . 00068 

 .00085 

 .00029 

 . 00052 

 .00068 

 .00017 

 . 00074 

 .00036 

 .00018 

 . 00057 

 .00017 



0. 0775 

 .0382 

 .0404 

 .0935 

 .0679 

 . 0399 

 . 0578 

 .0351 

 .0042 

 . 0067 

 . 0032 

 .0040 

 .5390 



Distinct blue color. 



Do. 



Do. 



Do. 

 Pale blue color. 



Do. 

 Distinct blue color. 



Do.< 

 Pale blue color. 

 Very pale l)lue color. 

 Pale blue color. 



Do. 



> Hillebrani, W. V.: Bull. 42?, U. s. '"Jeil. Surver, l!-in, -jig pp. 



2 Precipitate of cobalt sulphide was too small to handle gravimetrically but was ignited and taken 

 up in a borax bead with the color results shown in table. 



8 Same as II, except more deeply colored and a s'n^le larce crystal was used for analysis. 

 '■: * Bead was colored a deeper blue than for anv other of the 12 samples analyzed. 

 I 6 Analysis by Robert Robertson. The Virginias, 1885, vol. 6, p. 2. 



