284 Headden — Tunystous Oxide, new Oxide of Tungsten. 



This and also the preceding analysis indicates an excess of 

 tungsten over that required by the oxygen for the formation 

 of tungstous oxide. But if there remained any metallic tung- 

 sten in the regulus we would expect to find it here ; because 

 of its high specific gravity and' the fact that careful washing 

 was the only available method for separating and purifying the 

 portions into which the residues were divided. Boiling the 

 material on which analysis No. Ill was made with hydric nitrate 

 had no effect upon this point as columbous oxide and metallic 

 tungsten are either only very slowly attacked or not at all 

 acted upon by this agent. 



Relative to the probability of the presence of metallic tung- 

 sten in these iron bottoms I would repeat the statement that 

 they are closely related to cast iron and others have observed 

 that metallic tungsten may exist as such in cast iron. Further, 

 the result of much labor in an endeavor to obtain columbous 

 oxide directly from a mixture of columbite, magnetite, and 

 stream tin containing wolfram, was a series of iron bottoms 

 which when melted together and subsequently dissolved in 

 hydric chloride yielded a small quantity of a well crystallized 

 residue in which, neglecting a minute quantity of iron, tung- 

 sten was the only metal. The crystals were octahedral, appar- 

 ently quadratic, with some six-sided plates similar in color, 

 luster, and in being pitted to those observed in portions two 

 and three of the residue form the third bottom. Most of the 

 octahedral faces were smooth but on some crystals tuese faces 

 were depressed at the center. These crystals when kept at a 

 bright red heat for half an hour in an open crucible were only 

 very slightly oxidized and appear to be metallic tungsten.* 



As the components of these various products could not be 

 satisfactorily distinguished, even with the aid of the micro- 

 scope, we are compelled to have recourse to the study of those 

 ratios which remain constant while others vary. 



We may divide the results obtained in analyses one and two 

 into two groups, one containing the iron and tin, and the 

 other, the columbium, tungsten, and oxygen. The iron and 

 tin together constitute 25 per cent of the first sample and 

 rather less than 4 per cent of the second, but the ratio of the 

 iron to the tin is the same in each. If we calculate these ele- 

 ments found in each analysis to one hundred we obtain for 



* The ignited crystals themselves were not soluble in hot hydric chloride, sp. gr. 

 l - 2, but the ignited mass gave with this salt a yellowish solution which solidified 

 on being poured into a cold beaker, to a buttery mass, which proved to be pure 

 tungstic acid. This is the second instance in which I have found tungstic acid to 

 be soluble in hydric chloride. The statement in Gmelin-Kraut's Handbuch der 

 Chemie, is that ■' Acids, even concentrated sulphuric acid, do not dissolve it," i. e. 

 W0 3 . Its solubility in acids has been observed but once before this so far as I 

 can find, i. e. by J. W. Mallet. 



