4:40 Hess and Wells — Occurrence of Struverite. 



Residue 4*4 4-1 



TiO ..48-0 51-2 



Ta a O, ..340 33-0 



Cb a O, 2-6 4-3 



FeO 6-8 



These results were corrected slightly for the residue in the 

 boat. This residue was shown to be 50 per cent SiO a . The 

 remainder was computed as 90 per cent (Ta,Ch)„0 6 and 10 per 

 cent TiO a . With these corrections the results become : 



SiO a 2-2 2-0 



TiO a 48-2 51-4 



Ta a O B 36-6 34-6 



Cb a 6 2-8 4-5 



FeO 6-8 



An average of the results recorded on p. 436, with the results 

 above gives the final average in the first column below. On 

 microscopic evidence the SiO a is considered to be gangue. The 

 second column gives the composition of the anhydrous, gangue 

 free material. The third and fourth columns give the molecu- 

 lar ratios. 



H„0 0-4 



SiO a 2-0 



TiO a . 47-8 49-1 -613) 



SnO a 1-3 1-3 -009 \ b ^ 



FeO 7-3 7-5 -104 -104 



Ta„O t 34-8 35-7 -081 



Cb„0_. 6-2 6-4 -024 r ; 



99-8 100-0 



These results are sufficient to establish the essential composi- 

 tion of the mineral. In view of the fact that the determination 

 of columbium by the Metzger and Taylor method is undoubt- 

 edly better than by the first method used, it seems likely that 

 the tantalum content should be a little higher and the colum- 

 bium content a little lower than here stated, but the higher FeO 

 content, by the first method, 7*5 per cent, probably deserves 

 the greater weight. 



The mineral designated " black rutile " by Headden,* which 

 was probably this mineral, was stated by him to consist of TiO a 

 90-79 per cent, FeO 8-01, SnO a 1-35, MnO trace. If the tan- 

 talum and columbium oxides determined above are added to 

 the titanium oxide, the result is : Acid earths, 91-2, FeO 7*5, 

 SnO a 1'3, a striking confirmation of the suspicion, raised by the 

 density, that Headden weighed the tantalum and columbium 

 *This Journal [3], xli, 249, 1891. 



