106 



TIIK CRYSTAL FALLS IRON-BEARING DISTRICT. 



The value of such an investigation largely depends on the freshness of 

 the rocks examined and the amount of variation. The Hemlock volcanics 

 are all more or less altered, and the variation in character is slight. I wish, 

 however, to call attention to the close relationship exhibited by the types of 

 which analyses were made, and to that end the analysis of the nonpor- 

 phyritic very basic appearing basalt (No. 3 of Table I) is repeated and is 

 placed by the side of the analyses of the porphyritic ones. The complete 

 analyses made by Dr. Henry M. Stokes, of the United States Geological 

 Survey, are found in the first table. In the second table there is given the 

 molecular proportion of the chief oxides, those which are not here repre- 

 sented having been first proportioned among them. From this table, 

 following Rosenbusch,^ there were obtained the figures in the third table, 

 showing the atomic proportions of the metals present," 



The analyses are arranged according to the increasing percentage of 



calcium. 



Analyses of porphyritic metabasalt. 



TABLE I.— COMPLETE ANALYSES. 



CoDStitnont. 



SiOa 



TiO, 



AlgOn 



FeiO, 



FeO 



MnO 



CaO 



MgO 



Na^O 



KoO 



PsO. 



CI 



S03 



C02 



HjOat llO-^ .... 

 HsO above 110'^. 



Total 



47.20 

 3.30 



15.36 

 3.06 



8.87 

 .20 

 5.05 

 4.20 

 4.72 

 1.40 

 .36 



52.59' 



1.36 



15.93 



6.12 



3.96 



.25 



5.55 



5.04 



5.79 



.67 



.15 



3.34 



.16 



3.04 



None. 



.16 



2.16 



99.73 



r3. 



46.47 



1.28 



16.28 



3.15 



8.96 



.09 



7.90 



6.56 



3.64 



.21 



.13 



1.26 



.28 



3.89 



oNo. 3 is the analysis by Dr. Stokes of the nonporphyritlc basalt, and is given lor comparison. 



' Uber die chemischen Beziebungen der Eruptivgesteine, by H. Eosenbusch: Tsch. Min. u. Pet. 

 Mitt., Vol. XI, 1890, p. 144. 



^Tables No.s. II and III were calculated for me by Mr. Victor H. Bassett, assistant in cbcmistry 

 in the University of Wisconsin. 



