SCIENCE 



[Vol. XXI. No. 524 



It is not the theory of a science which urges the progress of that 

 science, but it is the attempt to discover whether or not the sug- 

 gested theory will explain the facts of the science, that leads to 

 the latter's rapid development. The suggestion of the atomic 

 theory demanded its discussion, and it was this discussion that 

 advanced chemistry to the position it now occupies among the 

 exact sciences. The theory of evolution did not by any means 

 explain away the difficulty of accounting for the existence of 

 many species of living things, but it was the attempt to discover 

 whether the theory is founded on a secure basis or not, that has 

 led to the wonderful progress of biology within the past quarter 

 of a century. So the mere suggestion of Mr. Iddings's theory as 

 to the origin of eruptive rocks, because of its comprehensiveness, 

 is bound to lead to discussion that will in the end give us a con- 

 ception of the cause of the almost infinite variety among these 

 bodies more simple than any other conception that has thu^ far 

 been held. 



Ml'. Iddings was highly favored in the beginning ' of his studies 

 by the opportunity afforded him of comparing the deep-seated 

 portions of a series of rocks with their surface equivalents. Elec- 

 tric Peak and Sepulchre Mountain, in the Yellowstone National 

 Park, are separated from each other by a great fault, in conse- 

 quence of which the intrusive stock and its apophyses of Electric 

 Peak are brought to the same horizon with the dykes and surface- 

 flows of Sepulchre Mountain. 



Upon compariu'.? the Electric Peak intrusives with the Sepul- 

 chre Mountiin effusives, it was found that, although each group 

 comprehends a complex series of rock-types, the two groups have, 

 on the whole, a striking similarity m composition. Certain 

 characteristic minerals found in the intrusives are also common 

 in the efifusives. Moreover, the transition between the members 

 of each series is so very gradual that it is impossible to draw any 

 sharp line between the different types. These facts indicate the 

 existence of a close relationship between the typical intrusives of 

 Electric Peak and the typical effusives of Sepulchre Mountain, 

 and a unity of origin for the members of each series, with a 

 gradual change in the conditions under which the different mem- 

 bers were formed. Though the ind ividual members of the effusives 

 differ markedly in structure from the members of the intrusive 

 group, the two groups are regarded as having resulted from the 

 cooling of what was originally one mass of magma, but which, 

 in consequence of a differentiatioo of its parts, became separated 

 into various magmas differing in composition. The differentiated 

 magmas, upon their extrusion from the depths, consolidated as 

 widely differing rocks, either of the intrusive type, or of the 

 effusive type, according as the magmas cooled beneath the sur- 

 face or upon it. 



Examination of other regions of eruptive rocks reveals the same 

 relationship existing between the various rock-types occurring in 

 them. There is a more or less striking similarity in some re- 

 spects between all types occurring within a region covered by 

 rocks extruded from a single centre, and a marked difference be- 

 tween these and the series of rocks of other regions. Thus the 

 rocks of a single eruptive centre are more closely related to each 

 other than to similar mineralogical aggregates originating at a 

 different centre, or, as Mr. Iddings expresses it. the rocks of a 

 single centre are consanguinous. 



No matter how different in mineralogical composition and in 

 structure, all the products of a given centre — consanguinous 

 products — should be grouped together in a classification of rocks, 

 rather than rocks of similar mineralogical composition and 

 similar structure from widely separated regions of volcanic ac- 

 tivity. The differences in structure and mineralogical composi- 

 tion of consanguinous rocks are the result of the differentiation 

 of the magma from which they were derived, together with dif- 

 ferences in the conditions under which the differentiated parts of 

 this magma were cooled. Their chemical peculiarities are the 

 direct result of the chemical nature of the homogeneous magma 

 before its differentiation into parts. If this notion is correct, the 

 succession of products originating during the course of a volcanic 

 extrusion should be "from a rock of average composition through 



= The Eruptive Kocka of Electric Peak and Sepulclire Mountain, Teilow- 

 stone National Park. 12th Ann. Eep. Director U. S. Geol. Survey, p. 569. 



siliceous and more siliceous ones to rocksTextremely low in silica 

 and others extremely high in silica, that is, the series commences 

 with a mean and ends with extremes." 



It will be the endeavor to discover whether this law of succes- 

 sion expresses the facts in the case or not, that will advance the 

 science of petrography to that of petrology. If Mr. Innings's law 

 of succession is found to hold, the future classification of rocks 

 will be based upon the principle of consanguinity: there will be 

 grouped in the same great division types of different mineralogical 

 composition and of different structure, while the different great 

 divisions will be based primarily upon chemical considerations. 

 What these chemical considerations are to be it is difficult at 

 present to foresee. 



Whatever may he the future classification of rocks, however, 

 it is quite certain that petrographers are in the main right in 

 distinguishing between rocks of different structures and different 

 mineralogical composition by different names. There is a fash- 

 ionable tendency apparent among English and American petrog- 

 raphers to decry the habit of naming these slight differences, not 

 because the number of rock-types in nature is in reality small, 

 but simply because the terminology of petrography by the addi- 

 tion of these names becomes large — as if we could increase the 

 simplicity of the science by refusing names to the objects of whose 

 study it consists. The same tendency has been observed also in 

 the history of chemistry. Some inorganic chemists have objected 

 seriously to the introduction of the many new terms into organic 

 chemistry, and yet nothing has done more to advance this par- 

 ticular phase of the science than its system of nomenclature. It 

 is easily understood why geologists should object to the increase 

 in rock names, since this increase necessitates a greater amount 

 of labor upon their part in becoming acquainted with the terms. 

 But why petrographers should object to a more accurate designa- 

 tion of the objects of their study is not understandable. It would 

 seem to the writer that for petrographical purposes every rock- 

 type that differs in some one essential feature from all other 

 rock-types should receive a distinctive name, in order that its 

 differences might be emphasized. If all the types with major 

 characteristics in common should be grouped under the same 

 name, we should lose sight entirely of their minor characteristics 

 that may be exceedingly important as throwing light on the rela- 

 tion of composition and structure to the conditions under which 

 the rocks were formed. Again, it is much more convenient to 

 speak of a keratophyre than of a "granophyric granite differing 

 from ordinary granophyre in the possession of anorthoclase instead 

 of orthoclase." This difference between keratophyre and grano- 

 phyre, though of insignificant importance from the point of view 

 of the geologist, ought to be of considerable importance to the 

 specialist in rocks. It may express simply a difference in the 

 original constitution of the magma from which the rock was 

 formed, or it may be the expression of peculiar conditions under 

 which solidification took place. In either case the difference is of 

 importance and should be emphasized. 



It would appear that the difficulty to the geologist of acquaint- 

 ing himself with the complete terminology of petrography might 

 be avoided by grouping rocks in accordance with their chemical 

 composition and structural swm7arrtjes, and by dividing the groups 

 according to the differences prevailing among their members. 

 Geologists need take account merely of the great groups, while 

 petrographers would require to become acquainted with their sub- 

 divisions. 



In denying the necessity of expressing in their names the com- 

 paratively slight differences noted between many rocks, it will 

 not do to say that petrography is simply a branch of geology and 

 that there is no room for the study apart from geology. The 

 methods of petrography are entirely different from those of geol- 

 ogy ; in many cases they are as different from those of the last- 

 named science as are those of paleontology. Petrography is the 

 special science dealing with some of the materials of geology. 

 Unless it is recognized as distinct from geology it will never be- 

 come of the importance that it will otherwise assume, and cannot 

 aid geology as it should do. If it be regarded as something 

 worthy of study for its own sake, then it is necessary to label the 

 objects of its study so that they may be handled conveniently. 



