EARTH'S CBUST — WASHINGTON. 285 



microscope the groundmass would show feldspars, pyroxene (or horn- 

 blende), and possibly a little quartz, with small grains of magnetite 

 and apatite, and with or without glass, according to the rapidity of 

 cooling. 



Stated in quantitative terms of " modal " or actual minerals, the 

 rocks would have probably the following approximate compositions : 



Grano- Ande- 



diorite. . site. 



Quartz 



Andesine (Na-Ca feldspar) . 



Orthoclase (K feldspar) 



Hornblende and Motile 



Pyroxene 



Magnetite 



Apatite 



11 10 



47 47 



16 I 18 

 20 \ 



5 | 5 



1 1 



It will be seen that, in either case, the average rock would be com- 

 posed entirely of the most common minerals, as is to be expected, with 

 the exceptions of olivine, nephelite, and leucite, which are much less 

 often met with and which, furthermore, are not found in rocks with 

 an excess of silica, as is true of the average rock. Inasmuch as the 

 average rock would have been formed at some depth beneath the sur- 

 face, the average crust may be considered to be a granodiorite, with 

 the general characters and mineralogical composition described 

 briefly above. The fact, however, must not be lost sight of that locally 

 the crustal rock may vary within very wide limits, as will be pointed 

 out later. We are dealing here only with the average of the crust 

 as a whole. 



We may examine the chemical composition of the earth's crust in 

 greater detail and, as has been done by Clarke in the papers cited 

 above, reduce the figures of the analysis to the form of the component 

 elements. The results are given in the annexed Table II, there being 

 here presented, not only the elementary constituents of the average 

 rock given in Table I, but in addition data showing the relative 

 abundance of some other of the more important elements that are not 

 usually, or indeed are never, determined in the analyses of rocks. The 

 data for these are taken from estimates by Vogt, De Launay, ami 

 Kemp, with'additional data by Clarke and Steiger for a few, and sonic 

 additions and changes in relative position based on my own studies. 

 An " x " means a digit in the respective decimal place or places. The 

 elements are presented in their order of relative abundance. This 

 estimate is to be regarded as provisional. 



This average, it must be repeated, does not include the sedimentary 

 rocks or constituents of the hydrosphere or of the atmosphere. Clarke 

 has included these in several of his estimates, and his latest shows 

 that the percentage, on this basis, of oxygen is 50.02, of silicon 25.80, 



