OBSERVATIONS IN CHEMISTRY AND MINERALOGY. 425 



For analysis 0.245 gram of pure material was available. Was used in hree 

 portions. 0.1 003 gave loss by ignit ion 0.0064 gram ; silica 0.0549 gram ; fcrrioxyd 

 and alumina 0.0270 gram; magnesium pyrophosphate 0.0160 gram. 



0.0413 gram was decomposed with sulfuric and hydrofluori ids in atmo- 

 sphere of carbon dioxyd. 0.3 c.c. permanganate solution were required. 



0.1003 gave potassium sulfate 0.0137 gram; this included a trace of sodium 



sul f at < 



SiO. - 54.73 : 60 - 0.9122 

 Fe,0, = 13.44 : 160 - 0.0840 ^ , _ UI 

 Al*Oi - 7.66 : 102.2 - 0.0741 J^- 1 * 1 



FeO - 5.30: 72 - 0.0736 1 



MgO - 5.76: 40 - 0.1440 L 

 CaO none r u 



K,0 - 7.40: 94 -0.0787 



11,0 - 6.40 : 18 - 0.3566 



100.59 



2963 



The ratio is very nearly RO : R 2 0, : SiO, : H,0 = 2:1:6: 2.24. Aouming 

 that one molecule of water is of basic character, the formula transforms into a 

 clear metasilicate 



(K 2 H 2 Mg, Fe),(Fe 2 Al 2 ) . (SiO,) 4 + 1.24(H,0). 



8. On Natrojarosite from New Locality. 



The material was collected during the summer of 1910 by Professor Seaman 

 at the Mina San Toy, Santa Eulalia, Chihuahua, Mexico; he placed it in my 



hands for investigation. 



The material appears in association with selenite and mimetite, as well 



tome one as yet not identified silver mineral, because the aggregate assays up to 

 60 ounces of silver to the ton. Under the lens the natrojarosite forma an aggre- 

 gate of straw-yellow scales, with strong silky luster. The scales are lying loosely 

 together and resemble closely the scaly dust on the wings of a butterfly. Like 

 the latter, they stick to the finger at the gentlest touch. Cacoxenite was th< 

 nearest mineral suggested to my memory. With one-half inch objective I hi 

 scales appear as well-defined hexagonal plates. The edges do not show any faces 

 The habitus is altogether holohedral. Herein they differ from those described 

 „ Hildebrand & Penfield 1 from the east side of Soda Spring Valley, Nevada. 

 The latter crystals are rhombohedral, same as the normal jarosite and alunitc 



B. B. Same reactions as those of normal jarosite. 



0.5000 gram gave 0.2547Fe 2 O 3 ; 0.4423BaSO 4 - 0.1517SO,; 0.056(K, Na)Cl; 



0.0525K,PtCl 



2 x tv^ie 



0.2554 gram mixed with PbO gave at red heat a loss of 0.0309 1 1 2 



Insoluble - 0.69 



Fe,0, - 50.94 : 160 - 0.3184 3.57 

 SO, - 30.43: 80 - 0.3804 4.263 



am 



NajO - 4.20 : 62 - 0.0677 

 K,0 - 2.02: 94-0.0215 



H *0 " 11.73 : 18 - 0.6517 7.3 



100.01 



Am. J. Science, XIV, fourth series, pp. 211 and following. 



