EARTH'S CKUST WASHINGTON. 301 



platinum in a region of sodic rocks, but would here look rather for 

 the minerals of cerium, the rare earths, uranium, or tungsten. 



COMAGMATIC REGIONS. 



Let us now return to the earth's crust and endeavor to answer the 

 second question propounded above, namely, whether all large por- 

 tions of the crust are alike in general, or whether they show marked 

 differences; that is, whether the crust is essentially alike or unlike 

 over different areas. 



Nearly 50 years ago Vogelsang 29 pointed out that the igneous rocks 

 of certain districts showed certain textural or mineral characters in 

 common, which served to distinguish them from the rocks of other 

 districts. The same idea was expressed later by Judd, 30 and still later 

 by Iddings, 31 the latter showing that the differences between different 

 districts were referable ultimately to differences in the chemical com- 

 position of the rocks. Such districts were called " geognostische 

 Bezirke" by Vogelsang, " petrographic provinces" by Judd, the 

 latter name being that in common use, Iddings using the term " con- 

 sanguinity," while the writer later 32 called them "comagmatic 

 regions," to indicate the idea that the various rocks of a given region 

 are derived from a common magma, by processes of so-called differ- 

 entiation. Into the discussion of differentiation we can not even 

 begin to enter here, though it forms one of the most important and 

 most complex features of petrology, the science of rocks. 



The proper study of petrographic provinces, or, as we shall here 

 term them, comagmatic regions, is as yet, so to speak, in its infancy. 

 Only a few regions have been described at all adequately from the 

 most general point of view, such as the Christiania region in south- 

 ern Norway, that of central Montana, the Yellowstone Park, and 

 the volcanoes of western Italy ; and these descriptions leave much to 

 be desired. 



Indeed, even the fundamental data for our definition of a comag- 

 matic region are somewhat uncertain and the application of the 

 idea is somewhat loose. Thus, considering the time element, the 

 life of a region may extend over many geological periods, as that 

 of Great Britain from the Silurian to the Tertiary; or it may be 

 confined to but little more than one period, as with the western 

 Italian volcanoes. The areal extent may vary from many thousands 

 of square miles to a few hundreds, though we are beginning to be- 

 lieve that the smaller " regions " are probably to be regarded as but 

 parts of larger ones. The shape of the area may also vary ; it may be 

 more or less equilateral, a long zone, either broad or narrow and 



"Vogelsang, H., Zeits. doutsch. geol. Ges., xxiv, p. 525, 1872. 

 »°Judd, J. W., Quart. Jour. Geol. Soc, xlii, p. 54, 1886. 

 *» Iddings, .1. P., Bull. Phil. Soc. Wash., xii, p. 128, 1892. 

 » Washington, H. S., Carnegie Inst. Publ., No. 57, p. 5, 1906. 



