﻿Yol. 66.] CLASSIFICATION OF IGNEOUS EOCKS. 475 



Classification by geographical distribution of 

 chemically different rocks. 



If the igneous rocks of the northern and southern hemispheres, 

 or of the several continental masses, differed radically in chemical 

 composition, with no or but few common types, we might use this 

 fact as a primary means of classification, even if the reason for such 

 differences were not understood. But the rocks exhibit no such 

 sharp limitations in occurrence, although certain notable features 

 of distribution have been pointed out. Some attempts have been 

 made to introduce these factors into petrographic system. 



The Atlantic and Pacific kindred. — In 1892 Iddings 

 pointed out that rocks rich in alkali seemed to be abundant in 

 certain parts of the world and rare in others, 1 outlining, to some 

 extent their distribution. Later, Barker commented on the corre- 

 spondence of the areas characterized by the alkali and subalkali 

 groups with the broad Atlantic and Pacific basins, and suggested a 

 kind of correlation between these magmatic facies and crustal 

 movements. 2 But this vague distinction of chemical groups first 

 assumed systematic importance through the proposition of Becke 3 

 to distinguish the Atlantic and Pacific ' Sippen ' or kindred, 

 which is, in fact, as he states, an attempt to provide brief and 

 convenient terms for cumbersome expressions used by Rosenbusch 

 and his followers to designate the ' foyaitic-theralitic ' and ' granito- 

 dioritic and gabbro-peridotitic ' magma series. These series are 

 really based upon chemical differences, and will be referred to 

 again. Aside from the question as to whether such a primary 

 chemical division is justifiable, the use of the terms Atlantic and 

 Pacific, in the manner suggested by Becke, implies a generalization 

 as to facts of distribution to which there is such a multitude of 

 exceptions that its expression in system is unwarranted. 



It may be true that the lands adjacent to the Atlantic Ocean 

 are richer in the so-called alkali-magmas than those bordering the 

 Pacific, but existing data concerning large areas involved in this 

 generalization are very meagre, and the question has never been 

 discussed in print with the thoroughness which it deserves. This 

 is manifestly not the place for extended comment on this matter, 

 but I wish to point out certain facts among many which might be 

 cited as bearing on the problem. 



Bohemia, the younger rocks of whicli country Becke strangely 

 selects as representing the Atlantic kindred, lies between Hungary 

 and Germany. The great series of volcanic rocks of the former 

 country are admittedly typical of the Pacific kindred ; and Germany, 

 though exhibiting many rocks of both series, is particularly the 

 land of the Pacific lamprophyres — minette, kersantite, vogesite, 



1 ' The Origin of Igneous Rocks ' Bull. Phil. Soc. Wash. vol. xii (1892-95) 

 pp. 89-213 (pp. 183, 184). 



2 : The Natural History of Igneous Rocks' Science Progress, vol. vi (1897) 

 pp. 12-33. 



:( ' Die Eruptivgebiete des Bohmisehen Mittelgebirges & der Aiuenkan- 

 ischen Andes' Tscherm. Min. Petr. Mitth. vol. xsii (1903) pp. 209-f,:,. 



•1 k l' 



