C. A. Joy on a Meteorite from Chili. 245 
As the analysis was conducted with great care; and as we 
have alumina and lime evidently derived from the decomposi- 
tion of a portion of the mineral, and as protoxyd of iron is 
easily attacked in silicates, it is proper to assume that the differ- 
ence is due to oxygen combined with iron as protoxyd. Assum- 
ing 1:90 pr. ct. oxygen, we require 6°65 pr. ct. Fe to form 8°55 
FeO. This will give us for the soluble portion : 
Fe 70°83 pr.ct. Mn 0°55 pr. ete 
Ni TT? Al,0, 5-02 
Co 1:23 FeO 8:55 
Cu 0-06 CaO 1°80 
S 3-95 SnO, 0-04 
P 017 gpg 
The average of several analyses gave 68°19 pr. ct. soluble in 
acids, and 31°81 pr. ct. insoluble in acids. 
Insoluble mineral portion. 
tion of HCl yielded brown sulphid of tin. The ammoniacal fil- 
trate from SnO, was colored distinctly blue by co 
The filtrate from the sulphid of tin and sulphid of copper was 
ted with chlorid of ammonium and ammonia and saturated 
_ With sulphid of ammonium and heated, by which iron, manga- 
Cipitate was dissolved in aqua regia. From the filtrate from the 
7 tg metals the lime was precipitated by oxalate of ammonia 
the magnesia by phosphate of soda. 
From the solution in aqua regia, the oxyds of iron, alumina, 
| mium, were precipitated by carbonate of baryta in the 
cold. The excess of baryta was removed by sulphuric acid and 
Manganese precipitated by potash: this precipitate was ex- 
amined for nickel and cobalt and found to contain traces. 
AM. Jour. Sct.—Seconp Serres, Vou. XXXVII, No. 110.—Mancx, 1864. 
32 
hese, nickel, cobalt, and chromium were ee aprta This pre- 
