112 /S. L. Pen field — Canfieldite, a new Germanium Mineral. 



antimony. As the mineral dissolves completely in nitric acid 

 tin cannot be present. These results therefore showed that the 

 germanium was satisfactorily pure. Another method of 

 analysis in which everything was determined in one portion is 

 as follows : Solution of the mineral in nitric acid, precipitation 

 of the silver with hydrochloric acid, of the sulphur with 

 barium nitrate, removal of the excess of chlorine and barium 

 in one operation with silver nitrate and sulphuric acid, iinal 

 removal of the silver by ammonium thiocyanate and determi- 

 nation of the germanium in the filtrate as above. 

 • The result of the analysis gave the following figures : 













Deducting 



Theory for 











Average. 



impurities. 



Ag 8 GeS 6 



s 



17-03 



17-04 





17-04 



17-10 



17-06 



Ge.-.- 



6-51 



6-52 



6-61 



6-55 



6-57 



6*42 



Ag. . 



76-01 



76-09 





76-05 



76-33 



76-52 



Fe, Zn 



•14 



•16 



•10 



•13 







Insol. _ 



•29 







•29 







100-06 100-00 100-00 



The formula of the mineral is evidently Ag 8 GeS 6 or 4Ag 2 S . 

 GeS 2 . The agreement of the analysis with the theory as will 

 be noticed is reasonably close. 



Winkler made the following analysis of argyrodite, from 

 which he derived the formula Ag 6 GeS 5 or 3Ag 2 S . GeS 2 . 



Analysis by 



Theory for 



Theory for 



Atomic 



Winkler. 



Ag 6 Ge 5 S 5 



Ag 8 GeS 6 



weights. 



S--- 17-13 



18-21 



17-00 



32 



Ge .. 6-93 



8-23 



6-42 



72-32 



Ag _. 74-72 



73-56 



76*52 



107-7 



Hg_. -31 









Fe _. -66 









Zn .. -22 









99-97 100-00 100-00 



It will be noticed that Winkler's analysis agrees much 

 more closely with the theory for Ag 8 GeS 6 , especially in 

 respect to the sulphur and germanium, than with the formula 

 advanced by him. It seems probable, therefore, that the two 

 minerals have the same chemical composition, but since Weis- 

 bach has shown that argyrodite is monoclinic and since can- 

 fieldite is isometric, they cannot be identical. 



In order to investigate this point more closely it seemed 

 desirable to make a new analysis of argyrodite by the same 

 methods which had been used for canfieldite. The material 

 was very carefully selected from an excellent specimen of the 



