GENESIS OF THE ALKALINE ROCKS 99 
magmas. That doubt was confirmed during observations in the 
volcanic fields of France, Italy, and Germany, in the colossal igneous 
fields of the North American Cordillera, and in Hawaii. 
North of the Hawaiian volcano, Hualalai, soda-rich trachyte 
rests on, and is surrounded by, great flows of olivine basalt. This 
spectacular association could not fail to recall the problem which 
had been so baffling during the interval of sixteen years. More 
than ever the writer was convinced of the extreme intimacy of the 
alkaline and subalkaline suites of igneous rocks. ‘The relatively 
small volume of all known alkaline rock and the small absolute 
sizes of alkaline bodies had suggested their derivation from the 
overwhelming subalkaline magmas, but it was not until the 
Hawaiian lavas were closely considered that a promising clue 
to an explanation was found. 
The hypothesis involving that clue was published in 1910. 
One outstanding fact on which the hypothesis is based is the com- 
mon occurrence of feldspathoids in the alkaline rocks instead of, 
or alongside, feldspars, which are the dominant constituents of the 
subalkaline rocks. The presence of nephelite or leucite signifies 
a lack of silica available for full saturation of soda or potash or 
both, suggesting some desilication of original subalkaline magma. 
Secondly, the concentration of alkalies in many alkaline rocks means 
that some agent or group of agents had collected the alkalies from 
the original magma. No success characterized the attempt to 
imagine adequate causes for the desilication or for the enrichment 
in alkalies, if the magma remained throughout purely juvenile 
in origin. The writer was thus led to assume the absorption of 
foreign material as the responsible condition. Such material is 
basic sedimentary rock. The most basic, large-scale, and widely 
spread rocks are limestone and dolomite. Also on account of the 
relative chemical simplicity of the carbonates the writer laid stress 
on these particular sediments. Unfortunately certain authors 
have thought that it was intended to explain all alkaline rocks by 
the interaction of resurgent carbonates with subalkaline magma, 
though that was disclaimed in the original paper. In 1913 the 
hypothesis was elaborated and the very important case of the 
t Bull. Geol. Soc. Am., XXI (1910), 108, 113, 114, 110. 
