ASSOCIATION OF ALKALINE AND SUBALKALINE ROCKS 89 



nephelite, leucite, comndum, etcetera; the extreme variability of alkaline 

 bodies in mineralogical and chemical composition; the occurrence of 

 such lime-bearing minerals as melilite, sea polite, wollastonite, melanite, 

 etcetera, and such COg-bearing minerals as cancrinite and (primary) cal- 

 cite. The difficult problem remains as to the physico-chemical reactions 

 involved in the h3^pothesis. This paper is offered as an advertisement of 

 the need for help from the physical chemists in giving a final explanation 

 of the described geological facts of rock-association. The writer is in- 

 debted to Dr. H. S. Washington for helpful, though adverse, criticism. 



Association of alkaline and subalkaline Eocks 



In most alkaline provinces typical basalts, diabases, gabbros, diorite, 

 andesite, (post-Archean) granite, or other common lime-alkali eruptives 

 occur. From the character and distribution of the visible pre-Cambrian 

 terrane, it is in the highest degree probable that its granite or greenstone 

 exists at greater or less depth in some part of each alkaline province on 

 the continental plateaus. This stead}^ association is partly illustrated in 

 the following table, which is far from being exhaustive even as regards the 

 post-Archean eruptives. Very often alkaline and subalkaline types are 

 represented in the same petrogenic cycle and province. A few examples 

 may be extracted from the table. The gabbro (with diorite) and diabase 

 of Ascutney Mountain, Vermont, respectively open and close the eruptive 

 series, which includes nordmarkite, paisanite, alkaline granite, campto- 

 nite, etcetera. The trachyte (acid phonolite) of Hawaii cuts, and is over- 

 flowed by, olivine basalt ; the nephelite basalt of Oahu is interbedded with 

 feldspar basalts. Feldspar basalts, with phonolites, compose Kerguelen, 

 Fernando I^oronha, Saint Helena, and other islands. 



Whether associated alkaline and subalkaline rocks are products of one 

 petrogenic cycle or have been erupted at long intervals of geological time, 

 it is hard, if not impossible, to conceive that the alkaline bodies are off- 

 shoots from primeval reservoirs of foyaitic or other magma rich in soda 

 and potash. This conception has held its place in petrology so long be- 

 cause of a systematic ignoring of the association described. The mind 

 fails to grasp the intercrustal or subcrustal conditions which permit of 

 the independent existence of initially alkaline and subalkaline magma 

 chambers beneath the same, generally quite limited, areas of the earth's 

 surface. 



The conception is further at fault in not allowing for the compara- 

 tively minute volume of all the alkaline-rock bodies now known on the 

 globe. The largest alkaline plutonic mass yet described is doubtless the 



