1890.] 



NEW YORK ACADEMY OF SCIENCES. 



203 



The approximate composition of the mass was got by extract- 

 ing everything possible by an electro-magnet, which took out 

 all the nickel iron and a little troilite, leaving the silicious part 

 and most of the troilite. Then the amount of S present in the 

 magnetic portion, and that in the silicious portion, was calcu- 

 lated as FeS ; the silicates were split into two portions by HCl, 

 and by the weights found in each case the given approximate 

 composition was calculated. Under the head of analysis of 

 nickeliferous iron is given the analysis of the metallic portion 

 after allowing for a very slight amount of attached silicates and 

 troilite. 



The analyses numbered from 1 to 5 are the residue left after re- 

 moving all the magnetic material. Column 1 is the part soluble in 

 HCl, column 4 that insoluble in HCl ; these two added together 

 would give the analysis as a whole of the non-magnetic portion. 

 Column 2 is the same analysis as 1, after removing the 2.67^ S 

 and an amount [6.01^^^] of FeO equivalent to the Fe necessary to 

 form troilite with the S. Column 3 is the same as 2 calculated 

 to 100^. Column 4, as stated, is the analysis of the insoluble 

 portion, and 5 is the same to 100^. It is of course probable 

 that the C\\0^ represents chromite, and possible that the alka- 

 lies and alumina with a little lime represent a soda-lime feld- 

 spar. 



The author takes great pleasure in thanking Mr. L. G. Eakins 

 and Prof. F. W. Clarke for their kindness, the former in fur- 

 nishing the analysis, and the latter in assisting with it. 



