76 



Penfield and Ford — Stibiotantalite. 



ordinates. Referring to page 73, the percentages of the total 

 acid oxides, (Nb,Taj 2 6 , of the three analyses have been 

 plotted on the vertical lines corresponding to the specific 

 gravities. The three points thus found do not fall quite on a 

 straight line, hence a circular arc is drawn through them and 

 continued. As given on page 74, the theoretical percentages of 

 Nb 2 5 and Ta 2 5 in (SbO) 2 Nb 2 6 and (SbO) 2 Ta 2 6 are, respec- 

 tively, 4:8-2 and 60*7 ; hence where the (Nb,Ta) 2 5 curve inter- 

 sects, horizontal lines corresponding to the above figures must 

 determine approximately the specific gravities of the two un- 

 known end products, namely 5*73 and 7*90. Next, the percent- 

 ages of Nb 2 6 and Ta 2 5 , as given on page 74, were plotted, and, 

 as shown by the figure, these fall very near two circular arcs, one 



























18 

























60 















































607 y ' 



























U 



J6J. 



ifeQ. 





















~<s 



50 



— 4 



f^ 















































, 































\ 











•<s 







1 





















I 



•KCfr 



9^ 



S 























2 







V 



























Nj 

















4 

 N 

































ts 











IS 









VS 









>. 



< 







:« 























v: 



?(1 





% 





£ 

















|& 













<* 











%' 







i 





















K 



^V 



%n 





















10 











/ 











































i 



/ 



/ 





























-^ 



















V 



/ 



















I 

























5-b 



6-0 



6-4 6-6 



Specific 



6-8 7-0 



Gravities. 



passing from percent to 48*2 per cent Nb 2 5 , the other from 

 per cent to 60*7 per cent Ta 2 5 . The greatest variation of 

 the determinations of JS r b 2 5 and Ta 2 5 from these curves is 

 not over one per cent, which is well within the errors of analysis. 

 If instead of using the values for .Nb 2 5 and Ta 2 5 given on 

 page 74, which were obtained indirectly from the percent- 

 ages of Sb 2 3 and Bi 2 3 and were selected because the three 

 analyses received like treatment, the values as given for our 

 two analyses on page 73 are plotted on figure 18, the points 

 fall almost exactly on the iN"b 2 5 and Ta 2 5 curves. Cer- 

 tainly, considering the difficulties of the analyses, the close 

 conformity of the determinations to the two curves is a most 

 satisfactory confirmation of the reliability of the results 

 obtained. It was by means of the diagram that the approxi- 

 mate composition of the two crystals studied optically, as 



