GRANULAR JMETAMOKPIIICS. 93 



Alnuiiua, Al-O' 12. Sli! 



Ferric oxide, Fe-O' l.2'j:i 



Ferrous oxide, FeO 3. 373 



Miinganous oxide, MuO 0.023 



Lime, CaO 1.823 



Magnesia, MgO 2.201) 



Soda, Na-0 fi. 033 



Potassa, K-O 1.259 



100. ilH 



Tlie utoriiic nitio represented by this aiuilysis is 11" : R" : K" : Sim 

 0.2G6 : 0.190 : 0.801 : 4.507. How the soda can be so greatly in excess 

 of tlie potash in sacli a rock I cannot exphxin. The cementing ftldspathic 

 mass may be very ricli in soda and possibly some of the unstriated feld- 

 spars are triclinic. It is also possible that the orthoclase is abnornjally sodic. 



These exaaqdes sliow that nearly all of the most important metaso- 

 matic processes can be traced in rocks the clastic character of which conld 

 be qnestioned by no one for a single moment. It is only necessary to sup- 

 pose the same piocesses carried further to obtain a product in which the 

 clastic character is obscured or obliterated, and the altered sandstones, 

 under the microscope no less than in the field, thus form ti-ansitions from 

 the clastic series to the holocrystalline rocks. The enlargement of clastic 

 grains by crystallization from infiltrating solutions, which has been shown 

 by several geologists ^ to be not infrequent in some regions, and of which 

 I too have studied fine examples, has not been observed in the rocks here 

 described. The general nature of the changes here consists in the attack of 

 clastic constituents, not in the addition of mineral matter of the same kind. 



GRANULAR METAMORPIUCS. 



Nomenclature. — Tlicro ai'O iu iiKiny parts of the world metamorphic rocks 

 very closely resembling diabase and diorite in mineral composition- 

 These rocks have sometimes been called, by myself as well as by others, 

 metamorphic diabases and metamorphic diorites; but there are serious ob- 

 jections to these terms. A diabase would be defined by most geologists as 

 a Pre-Tertiary eruptive rock, mainly com})osed of plagioclase and pyrox- 

 ene. This is also the historical meaning of the term. To those who have 



1 See Hunt, Origin of Crystalline Rocks, § 116. 



