72 PHOCEEDINGS OF THE NA TIONAL MUSEUM. 



supposed that the iron was fairly uniform. It is true that thTm^ 

 itself is large, yet these samples were taken from portions whose dis 

 tances from each other were so small that they could not be over 30 

 centimeters apart. 



The conclusion seems to lie ol)vious that a bulk analysis is of little 

 value unless made on the entire mass, or upon a very large sample, 

 conditions which in either case are practically impossible to comply 

 with. Second, that mineralog-ical separations of as large amounts as 

 possible, together with the study of sections, are the safest guide to the 

 composition of a meteoric mass. 



THE MINERALOGICAL COMPOSITION. 



In order to arrive at the chemical and physical char cters of the 

 mineral constituents, as troilite, schreibersite, etc., separations look- 

 ing to the isolation of these minerals were made. 



TROILITE. 



A nodule of troilite weighing I.52il grams wa? carefully picked 

 from the mass. This had a specific gravity of 4.789, a brass yellow 

 to bronze color, hardness about that of fluorite (4 in the scale), slightly 

 magnetic, and gave on analysis: 



Per cent. 



Jj'oi^ 63.40 



Nickel 20 



'"^"Ipli"'- 36. 21 



99.81 

 The mineral occurs somewhat sparingly through the mass, usually 

 in nodules varying considerably in size and shape, the one ranging 

 from 1 or 2 centimeters to a millimeter, the other from ne^arly 

 spherical to lenticular masses, and occasionally tilling veins, as shown 

 in the plate. So far as observed the segregation oi the sulphide is 

 usually accompanied by the segregation of graphite, commonly as a 

 very thin layer between the troilite and the iron ground mass. In no 

 in.stance was there noticed a segregation of the phosphide immediately 

 adjacent to the troilite, similar to conditions which may be observed 

 in certain sections of the Canyon Diablo and other irons. 



SCHREIBERSITE. 



Successive portions of the iron, amounting in all to 150 grams, were 

 treated with a large excess of hydrochloric acid (1 HCl, sp. g. 1.1 + 15 

 H,0). The several solutions were filtered, and the residues collected 

 and dried. The phosphide was then separated by a magnet, and 1.21 

 grams of small, brilliant steel-gray folia were thus collected. These 

 were, without further examination, treated with nitric acid for 

 analysis. It was noticed that while a portion of the material went 



