394 N- L. BOW EN 



given only to crystal settling. Since the publication of that paper a 

 number of writers have pointed out various features of igneous 

 rocks for which crystallization-differentiation has seemed to them 

 to offer no adequate explanation. These objections have arisen 

 largely from a failure to appreciate the importance to differentia- 

 tion of the period when igneous material is largely crystalline, that 

 is, the period when it is magma in that sense of the term to which 

 attention is called above. 



In the present paper the writer wishes to offer suggestions 

 regarding some of the more general objections that have been 

 raised against the theory of crystallization-differentiation and 

 especially to apply it to certain phenomena for whose explanation 

 it has been considered inadequate. 



RELATIVE IMPORTANCE OF SYNTEXIS 



In a recent paper on the " Genesis of the Alkaline Rocks" Daly 

 discusses many features of the genesis of igneous rocks in generaP 

 and takes the opportunity of criticizing certain aspects of the 

 theory of crystallization-differentiation. In his introduction and 

 at other points in the discussion Daly makes statements that 

 convey the impression that that theory, as a whole, must stand or 

 fall with its ability to explain the alkaline rocks. A brief examina- 

 tion of the basis on which the theory rests will show, however, that 

 such statements are hardly justified. From the results of a series 

 of experiments that dealt with the crystallization of silicate melts 

 the writer was led to the opinion that igneous rock types as we know 

 them in the field could be developed as a result of the fractional 

 crystallization of basic magma under appropriate conditions. The 

 actual experimental work was carried on with mixtures of simpler 

 constitution than the usual igneous rock, but frequently a fairly 

 close approach to the composition of an igneous rock in its more 

 essential aspects was accomplished. In such instances the con- 

 clusions drawn from the experimental work may be considered to 

 be of correspondingly greater rehability. If we examine the experi- 

 mental data we find that such close approach was made only in 

 the case of the subalkaline rocks, and in the development of the 



^Jour. GeoL, XXVI (1918), 97-134. 



