NO. 1273. A NEW METEORITE FROM KANSAS— MERRILL. 911 
gram consisted of small black grains, later identified as chromite, the 
soluble portions separated from which amounted to 3.2509 grams. 
From this the following percentages were obtained: 
Per cent. 
Fe 93 
Xi 6 
Co 0. 02 
S 0.03 
P 0. 025 
Cu Trace. 
This corresponds to: 
Nickeliferous iron ( Fe, Ni, Co, Cu, etc. ) 98. 273 
Schreibersite (Fe, Wi)^V' 1.645 
Troilite, FeS 0. 082 
100. 00 
The chromite obtained from various portions as a result of mineral- 
ogical separations was purified so far as possible bj^ treatment with 
acids, and 1.025 grams taken for anah^sis. This gave results as follows: 
Per cent. 
Cr^Og 65. 49 
Feb 33. 00 
MgC) 0. 40 
SiO, 0. 50 
It is a fair assumption that the magnesia and silica here shown 
belong to included olivine. 
A fresh sample, weighing 2.05 grams, yielded 0.028 per cent of 
chlorine, equivalent to 0.05 per cent of lawrencite. No traces of tin, 
platinum, or manganese could be detected. 
Specific gravity determinations made on masses weighing from 117.5 
to 139.35 grams, were found to var}'- from 3.95 to 4.2. The iron, it 
should be noted, was quite :ictive, taking quickly a coating of metallic 
copper when treated to a solution of copper sulphate. 
I can but feel that a great deal of importance may be attached to the 
peculiar structural features shown by this meteorite and the associa- 
tion of the metallic constituents. These latter have been plainly intro- 
duced since the first consolidation, and subsequent to the shattering of 
the olivines. They occupy the position of a binding constituent in a 
siliceous breccia. 
The source of the metallic constituents of meteorites has long been 
a matter of speculation, though it has been suggested by various 
authorities that it might result from the reduction of an iron-rich 
silicate. There is nothing in the present case to suggest any such 
origin, and it would seem to the writer much more probable that it 
should come from the lawrencite and troilite. That a portion of it 
thus resulted, is, it seems to me, extremely probable, from the con- 
ditions shown in Plate LV, above noted, in which we have the acicular 
