Headden — Tungstous Oxide, new Oxide of Tungsten. 285 



number I : Fe 80'47 per cent, Sn 19 "53 per cent, for number 

 II : Fe 81-72 per cent, Sn 1.8*28 per cent. The ratios ob- 

 tained for the remaining elements are quite as satisfactory and 

 represented in the same manner are 'as follows: 



No. I. \ T o. II. 



Cb 49-9L Cb 50-44 



W 38-42 W 39-09 



O 11-67 10-47 



100-00 100-00 



The percentage composition required by the mixture 

 10CbO+4WO is Cb 49*473, W 38-737, O 11-890. The prop- 

 erties of this mixture agree closely with those of columbous 

 oxide especially in burning with incandescence to columbic 

 acid. The presence of the triangular prisms is quite conclu- 

 sive that the columbium and tungsten are present in separate 

 compounds, which will appear more fully further on. 



The two remaining analyses are not comparable to the 

 same extent as those just given, still Nos. Ill and IV give 

 1 : 1-21 and 1 : 1-22, respectively for the ratios of the colum- 

 bium and tungsten to the oxygen. 



"While it is inferable from the results of analysis III that 

 the tungstic acid in the other analyses was derived from the 

 prismatic crystals, the most conclusive proof that these trian- 

 gular prisms are crystals of a compound of tungsten with only 

 a very little or no iron or tin is afforded by the material desig- 

 nated part I of the residue from the third bottom ; for upon 

 being treated with aqua regia in successive portions it yielded 

 upon the first treatment a solution carrying ferric oxide equal 

 to 2*3 times the amount of tungstic acid formed while the sec- 

 ond treatment produced an amount of tungstic acid equal to 

 7*3 times the amount of ferric oxide which went into solution. 

 The residue from the first treatment was very largely made 

 up of these triangular crystals. We conclude from all of this 

 that this hemiprism is the usual crystal form of the compound 

 of tungsten which occurs in these residues. This conclusion 

 is supported also by a comparison of the results obtained in 

 analyses II, III and IY with the respective materials on which 

 they were made. The material serving for analysis No. II 

 was a fine crystalline powder with some triangular crystals ; 

 that serving for No. Ill contained very many of them, while 

 in No. IY I estimated them as forming upwards of 85 per 

 cent of the mass. The results of the analyses are 



II. HI. IV. 



Cb, 48*56 per cent. 6 - 56 per cent. 1*46 per cent. 



W 37-64 " 76-12 " 86-64 " 



