700 DEPARTMENT OF THE INTERIOR 



2 GEORGE V., A. 1912 



of post-Archean differentiation, since magmatic differentiation is very probably 

 an incident of cooling nearly to the solidification point. 



Yet more telling is the argument that, if the basalt of the greater lava 

 floods is the product of a differentiation nearly contemporaneous with extrusion, 

 we should expect to find the other pole of the differentiation in immediate 

 association. This pole must be more acid than basalt, for no known earth-magma 

 can be fairly suggested which would, by splitting, give basalt as the acid pole. 

 The commoner peridotites are probably differentiates of basaltic magma and, in 

 any case, cannot be regarded as the parent of basalts. If, then, basalt is the basic 

 pole of magmatic differentiation we should expect to find large effusions of the 

 contemporaneous, more acid differentiate in the greater lava fields of the globe. 

 The more acid differentiate should normally overlie the basalt in the magma 

 chamber, and must in most cases be erupted through the opening fissures before 

 the basalt could reach the surface. The only escape from that conclusion is 

 to be found in the postulate that the acid differentiate had completely solidified 

 before the basalt was poured out. This postulate is plainly inconsistent with 

 geological experience in the smaller volcanic terranes, where both poles of mag- 

 matic differentiation are so regularly found in the extruded lavas. Yet in the 

 Columbia and Snake River lava fields of America, in the similarly vast field of 

 the Deccan, as in the ancient field covered by the Purcell Lava, there is no 

 acid differentiate to match the basic differentiate, basalt, of any of the fissure 

 eruptions. The simple and probable conclusion is that the basalt of all the vast 

 lava fields is pure, undifferentiated material from the earth's interior. 



If so, it seems to follow that no different kind of fluid rock-matter overlies 

 the basalt of the substratum. If, for example, a primary liparitic magma overlies 

 it, the method of eruption of the pure basalt through the liparite to the earth's 

 surface would be, to say the least, inconceivable. 



2. The association of chemical types at ' central eruptions ' (volcanic cones 

 and craters), is generally much more complex than that characteristic of the 

 greater lava plateaus. Two of the principal reasons for this are apparent. As 

 compared with the feeding sheets of lava in fissure eruptions, the lava columns of 

 cones stand longer in their vents. The vents of central eruptions are kept open 

 because of the emanation of gases from the feeding magma chamber. At the actual 

 opening the lava is sometimes seen to be superheated (e.g., at Hawaii, 

 Savaii, etc.) A moderate assimilation of the walls of the vent is to be 

 expected in the earlier stage of a volcano's history. Syntectics may be 

 formed; the primary magma may be subject to specially marked differentia- 

 tion through fluxing, or because of inoculation; and the syntectic may be differ- 

 entiated. Among so many possibilities it is little wonder that the sequence of 

 eruptive types is a variable one. After a vent has long afforded passage to lava, 

 so as to build up a first-class cone, the volcano approaches its limital height and 

 also the stage of extinction. Assimilation is then checked, the formerly enlarged 

 vent is narrowed by gradual freezing, and the final extrusions are composed of 

 primary magma or of its own differentiates. This appears to be the best explana- 

 tion of the general fact that the latest lavas of the largest volcanoes, like Etna 

 or Chimborazo, are basalts or pyroxene andesites. As noted below, there are 



