FIRST HYPOTHESIS OF DERIVATION 



447 



blende schist (IV), one may have been derived from aluminous clays or 

 shales rich in iron oxides, lime, and magnesia. Thus Dana accounts 

 for part of the pyroxenite and hornblendite beds inclosed in the lime- 

 stones of Westchester and New York counties : 



" This association of the limestone with rocks containing much black mica or 

 hornblende is, in fact, association with rocks containing much iron, as is indicated 

 by the rusting tendenc_v of the schists. This quality of these metamorphic schists 

 is a consequence of the ferruginous character of the original sedimentary beds 

 underlying or overlying the limestone strata. The iron of those sediments went, 

 for the most part, at the time of metamorphism, to make the black iron-bearing 

 mica or hornblende, the rest of it entering mainly into pyrite ,and sometimes 

 garnet."* 



A series of hornblende schists derived from such sediments has been 

 analyzed by W. H. Melville : f 



V. Pseudo-diorite, derived from arkose. Knoxville, California. 

 VI. Glaucophane schist, derived from shale. Sulphur Bank, California. 

 VII. Glaucophane schist, derived from shale. Monte Diablo, California. 





V. 



VI. 



VII. 



SiOj 



50.44 



49.68 



1.31 



13.60 



47 84 



TiOo 





ALOo 



8.18 



.48 

 1.06 

 6.29 



.21 

 11.55 

 17.63 



.50 

 2.98 



.98 



16.88 



Cro' >■> 





Fe.O, 



1.86 

 8.61 



.04 



10.96 



6.26 



.12 

 3.09 

 3.84 



.21 



4.99 



FeO. 



5.56 



MnO ... . . 



56 



CaO 



11 15 



MffO , ... 



7 89 



K2O 



NajO 



.46 

 3 20 



H2O 



1.98 



P„0« 



14 











100.30 



99.58 



100.65 



But by comparison of these with analysis IV it appears that the horn- 

 blende schist of Manhattan, with its low percentage of silica and high 

 content of iron oxides, mostly in ferrous condition, shows little relation 

 to any of the above schists ; nor, indeed, does that analysis approach 

 those of unaltered clays or shales, since in the latter the content of silica 

 usually much exceeds 55 per cent. In a highly basic shale from Clinton, 

 Indiana, the amount of silica is but 43.13 per cent, but with alumina 40.87 

 and ferric oxide onl}^ 3.44. 



* Am. Jour. Sci., vol. xx, 1880, p. 26. 



t Bull. U. S. Geol. Survey, no. 148, 1897, p. 222. 



