Page:The Journal of geology (IA journalofgeology11893univers).pdf/424

In South America the great Cordilleran system of the Andes presents a petrographical province identical, chemically and mineralogically, with those of the North American Cordilleras, and which appears to extend throughout its entire length. In the eastern part of the continent and on the islands off its coast the petrographical province is in turn identical in many respects with the eastern province of North America; the correspondence being most pronounced between the rocks from Brazil, described by Derby^[1], and those from Arkansas described by J. Francis Williams.^[2]

The chemical and mineralogical qualities or peculiarities which characterize the rocks of particular groups, and at the same time serve to distinguish them from those of some other groups, are like family traits of character, and suggest the intimate relationship and common origin of all of the igneous rocks of the group. They prove conclusively that the varieties of rocks occurring at a particular center of eruption, or in a volcanic district, have been derived from some magma common to the district by a process of differentiation similar to that which has caused smaller bodies of molten magma to become chemically heterogeneous and has produced mineralogical facies.

That the process which has produced the many kinds of igneous rocks in any region, with all their transitions into one another, was a process of differentiation of an originally homogeneous magma, and not the compounding of two or more different ones, is shown by the geological relationships between the various bodies of rock belonging to a volcanic center; more especially the order in which they have been erupted. A process dependent upon any set of physical conditions, which continues active for long periods of time must yield results that are to a very considerable extent functions of time, that is, they must be