802 H. S. WASHINGTON DECCAN SALTS AND PLATEAU BASALTS 



of the Thulean and Siberian regions, and by implication those of the 

 Dec-can, Oregonian, Algonkian, and Palisadan regions, should be consid- 

 ered as representing a distinct group, which he calls the Arctic, of the 

 same general classificatory rank as the Atlantic and the Pacific rocks. 

 With this suggestion von Wolff unconsciously shatters the dualistic con- 

 ception of igneous rocks, which comes down to us from Bunseirs study 

 of the Iceland lavas, and from even farther back in the early human 

 tendency to a dualistic interpretation of nature, as seen in night and day, 

 good and evil, Ormuzd and Ahriman. In studying igneous rocks we can 

 not limit ourselves to dualism; but, even assuming a possibly monistic 

 origin in the primordial magma, Ave must turn to a polyistic concept of 

 their relations. We must consider rocks as derived from or referable to 

 not only two lines of descent, but as coming down to us either from a 

 heterogeneous, diversified, or from a possibly single homogeneous earth 

 magma along many lines which may, and we have reason for thinking 

 do, intertwine or even throw back atavistically in their course of physico- 

 chemical evolution, one phase of which we call differentiation. On a 

 small scale this diverse evolution is seen in the glass and the holocrystal- 

 line parts of, let us say, the Deccan basalts or in the border and the 

 interior of a composite dike; on a large scale it is seen in the different 

 comagmatic regions or, still more broadly, in the petrologic difference 

 between the continental masses and the ocean floors. 



This last may seem to lead us inevitably to the usual dualistic idea of 

 the division of a primordial magma along two lines, ending in the two ex- 

 tremes of a "basaltic" and a "granitic" zone or stratum, according to the 

 theory of so-called gravitative adjustment, as first propounded by Darwin 

 and later by Daly and others. In reality, however, the course of differ- 

 entiation would seem to be not so simple, but to have given rise to various 

 chemically and consequently minerally quite diverse side lines, much as 

 we see to have been the case in organic evolution. We meet with an 

 overwhelming preponderance of granite and rhyolite, diorite and andesite, 

 gabbro and basalt, with all their various varieties, subvarieties, and 

 shades of subvarieties. But, large as nfay loom these common, universal 

 rocks, we must not forget the rarce aves of petrography, such as the ilmen- 

 ite-apatite rock nelsonite, the albite-rutile kragerite, or the magnetite 

 basalt arapahite, which, for the study of rock genesis, rock evolution, or 

 rock classification, may be of importance equal to the vastly greater and 

 more numerous masses of the familiar rocks mentioned above ; much as 

 the comparatively rare Peripatus and the Cycadaceae are of some special 

 interest in the study of the phylogeny of animals and plants. 



