224 JOSEPH P. IDDINGS 



amount of silica in the magma. It frequently occurs in associa- 

 tion with quartz or tridymite. It is a general law that the ferro- 

 magnesian minerals become more abundant as the quartz and 

 feldspathic minerals become less abundant. 



A method of correlating the chemical and mineral composi- 

 tion of igneous rocks on a basis of the silica percentage and of 

 the relative amounts of alkali in each case has been presented in 

 the article in this volume r already cited. It is now proposed, 

 without repeating what has been stated in that connection, to 

 discuss in greater detail some of the relations between the min- 

 eral and chemical constituents of igneous rocks by making use 

 of similar diagrams. In order to do this it will be necessary to 

 start with the simplest assumptions, and those that appear to 

 rest on the more general laws. Having obtained some of the 

 simpler conceptions, more complex ones may be attempted with 

 less danger of confusion. 



It is possible to obtain an idea of the range of quartz and of 

 leucite, nephelite and sodalite in igneous rocks by considering 

 the most favorable cases for their occurrence, and afterwards the 

 conditions that would modify their production. In order to 

 render the discussion as simple as possible it is advisable to con- 

 sider the extreme cases in which the alkali is either wholly soda 

 or wholly potash. In Diagram I the black spots represent the 

 rocks in which the soda is more than twice as great as potash, 

 and the red spots those in which the potash is greater than soda. 

 They correspond to the spots on Diagrams 2 and 4, Plates I and 

 II of this volume. They are introduced to aid the imagination. 

 The significance of the lines is explained in the text. For the 

 phrase " the analysis or analyses of rocks " in the discussion of 

 the diagrams the phrase "the rock or the rocks," will be substi- 

 tuted, it being understood that only the chemical attributes of 

 the rocks are under discussion. 



I. Let us first postulate the case of magmas whose alkali is 

 ivholly soda, and in which the molecules of alumina are always in 

 equal proportio7is to those of soda. Then with no other constituent 



1 Loc. cit., pp. 98-103. 



