10 



MEMOIRS NATIONAL ACADEMY OF SCIENCES. 



(Vol. xrv. 



more in diameter. In the selection of samples for analyses these were avoided as far as possible. 

 Carbon is present in form of microscopic diamonds and also as graphite. It has been the 

 subject of nimierous analyses, yielding variable restdts owing to its coarse crystallization. 

 Whitfield's analysis is given in column I below. In column II is given the average of three 

 analyses by Moissan, Booth, Gari-ett & Blair, and Tassin. 



Constituents . 



Silicon 



Sulphur 



Phosphorus 



Manganese 



Copper 



Nickel 



Cobalt 



Combined carbon . 

 Graph it ic carbon. . 



Iron oxides 



Iron protochloride 

 Iron 



Total 



Per cent. 



Trace. 



0.009 



.261 



None. 



.015 



7.335 



.510 



.105 



.028 



2.520 



.097 



S9. 167 



100.047 



Per cent. 



0.032 



.007 



.159 



None. 



Trace. 



5.828 



.044 



.465 



In process of analysis schreibersite to the amount of 1.832 per cent separated out. This 

 afforded the following composition: 



• Per cent. 



Iron 55. 04 



Nickel 29.58 



Phosphorus 15. 38 



Platinum was looked for in two portions of fifty grams each, but none was found. Neither is its 

 presence recorded in previous analyses by this same firm, by Derby," Tassin or Moissan.* J. W. 

 Mallet, working on a residue from the solution of 25 pounds of the iron in dilute hydrochloric 

 acid obtained residts representing 3.63 grams of platinum and 14.95 grams of iridiimi per metric 

 ton of the original iron, "with probably a trace of rhodium." He suggests, and this is in accord- 

 ance with our own results, that the platinoid metals are not uniformly distributed in the iron." 

 This may account for the failure to find it on the part of others. The quantity of material 

 worked upon is imdoubtedly an important factor, however. 



(2) Iron. — Casas Grandes, Mexico. Medium octahedrite. Previously analyzed and de- 

 scribed by Tassin."* Tlie results of Dr. Whitfield's analyses are given in Column I below. In 

 Coliunn II are given Mr. Tassin's results as previously obtained. 



Whitfield's analyses were made on 50-gram samples, free from evident inclusions of troilite. 

 The following elements were especially looked for but not found; antimony, arsenic, tin, lead, 

 palladium, platinum, titanium, tungsten, vanadium, uranium, chromium, manganese, molyb- 

 denimi, and zinc. 



a Amer. Joum. Sci., vol. 49, 1895, pp. 101-110. 



b See contributions to the Study of the Canyon Diablo Meteorite, by G. P. Merrill and W. Tassin, Smithson 

 50, pt. 2, 1907, p. 209. 



c Proc. Acad. Nat. Sci. Phila., Dec., 1905, p. 913. Footnote on p. 862. 

 (t Proc. U. S. Nat. Mus., vol. 25, 1902, pp. 69-74. 



I Misc. Coll., Quar. Issue, vol. 



