298 FirLp CoLuMpIAN MusEuM—GEOoLOoGY, VoL. I. 
although it has not infrequently been reported by other analysts as a 
constituent of the soluble portion of meteorites. 
Grouping the compounds of the above analysis which are known 
to enter into the composition of nickel-iron, chrysolite and bronzite, 
the following may be deduced as the probable composition of these 
three ingredients : 
COMPOSITION OF NICKEL-IRON. 

Per Cent. 
Fe : : ; 5° TOoOs 
Ni. : : . 20.26 
Co. . ; : a BSD Se ple e) 
100.00 
COMPOSITION OF SOLUBLE SILICATES (CHIEFLY CHRYSOLITE). 

Per Cent. 
ICE : ; ; : : ; : ; ‘ RO nae 
MgO . : ; ; : ; : ay ite SON 
FeO... t ; : ; : : : , : Shea ik 
Als O-F 2 : : : pe O<c04 
Ci On, ‘ . , BE OE: 
CaCl eae : : : ‘ : : d he Oe 
100.00 
Ratio of 2RO:SiO, : : 1.9997 : I. 
COMPOSITION OF INSOLUBLE SILICATES (CHIEFLY BRONZITE AND 
MONOCLINIC PYROXENES). 

Per Cea 
SRO Pee us ese e! Sah st Ce an BAe nt ch eee aaa Gaia 
Als Qay: : ; ' p eRe 
ee) ! ; eekat 8181 
MgO . : ; : ; : -2 S20 ee 
CaO ee ’ : : : th SuOM 
COs ve ; ; ; , : ; > Open 
INTO : : eh RE 
WajO're : : Fah Te : +f Repu 
Ketan : ; ‘ 3 : : : DCO WS, 5 
100.00 
Ratio of RO: SiO, : : 1:1.0148. 
To place the alkalies in the pyroxenes, as is here done, is con- 
trary to the usual custom, it being common to assume that they are 
present as feldspars. But as no feldspars could be detected in the 
‘slides and as alkalies are known to enter into the composition of 
pyroxenes in small amount, the conclusion here adopted seems the 
more reasonable one. No attempt was made to differentiate the two 
pyroxenes chemically, as I know of no guide for this. The amount 
of monoclinic pyroxene which can be seen in sections is very small, 
