March 24, 1887] 



NATURE 



483 



ties where the rocks are found, such as tonalite, 

 ortlerite, or palatinite ; or, lastly, following in the wake 

 of the organic chemists, we construct complex names by 

 stringing together those of the component minerals, as, 

 for example, quartz-augite-diorite or hornblencle-augite- 

 mica-andesite. The result is that either different petro- 

 graphers call the same rock by different names, or use 

 the same name to designate different rocks. At the same 

 time, so many views have arisen as to the fundamental 

 elements of petrographical classification, that there are 

 almost as many systems of classification as there are 

 petrographers. 



Up to the present time Continental geologists have 

 been in the habit of making geological age a primary 

 factor in classification. Xow, although this may apply to 

 Germany, it certainly will not hold for other countries. 

 In England and America it has been shown conclusively 

 that rocks identical in structure and composition have 

 been formed in pre-Tertiary and Tertiary times. 



The new edition of Rosenbusch's work would, it was 

 hoped, bring order into this chaos, and give us a classifi- 

 cation and nomenclature which, without being too rigid, 

 would allow of referring any particular rock to its family. 

 Such a classification, agreeing with all the known facts, 

 would doubtless readily be accepted by all geologists in 

 this country, were it only for the sake of uniformity and 

 unanimity. 



A brief review, or epitome, of Prcf. Rosenbusch's book 

 may not be unwelcome here. Any criticism had best be 

 reserved until the work is completed. We owe, indeed, 

 such a debt of gratitude to the author for collecting, 

 collating, and arranging the vast quantity of facts which 

 have been accumulating within the last few years, that it 

 would be almost presumptuous to attempt to find fault 

 with a work so excellent, so invaluable in every way. 



In the introduction Prof. Rosenbusch gives us his views 

 on classification. These differ very materially from those 

 expressed in the former edition. " A natural system of 

 classification must," he writes, " in the first place lay 

 stress on the geological mode of occurrence igcologische 

 Erschcinungsfonn), as determining structure and the 

 mineral components. In the second place comes chemical 

 composition, and, lastly, geological age." Secondary alter- 

 ation in structure or mineralogical composition can have 

 no classificatory value. 



It is the geological mode of occurrence that almost exclu- 

 sively detennines the structure of an eruptive rock. Erup- 

 ti\'e masses of the same chemical and mineralogical com- 

 position possess a totally different structure, according as 

 they were poured out at the earth's surface in the form of 

 lava, or consolidated in the deeper regions of the earth's 

 solid crust. This may, of course, also be expressed by 

 saying that the structure of a rock depends, ccctcris 

 paribus, only on the differences of temperature and 

 pressure to which it has been subjected during its 

 formation. 



Classified, then, according to their mode of occurrence, 

 eruptive rocks may be divided into two great groups : 

 (i) the Plutonic rocks {Tic/eiigesfcine) ; and (2) the 

 Volcanic or effusive rocks {Ergussgesteine). Occupying 

 an intermediate position between these two chief groups 

 is a third— that of the rocks occurring in the form of 

 dykes {Ganggesteinc). Both plutonic and volcanic rocks 



are^ often found \ as dykes ; but this group comprises 

 those rocks which are found occurring alone in this 

 form, and which possess certain struftural peculiarities 

 entitling them to be considered apart from the plutonic 

 and effusive rocks. With regard to the latter it may be 

 remarked that it is in this group that Prof. Rosenbusch 

 does not feel justified in dropping altogether geological 

 age as a classificatory factor. Accordingly these rocks 

 are subdivided by him into palaovolcatiic and neo- 

 volcanic ; the former embracing those erupted in pre- 

 Tertiary times, the latter those of Tertiary age. 



In the introductory chapter to the plutonic rocks Prof. 

 Rosenbusch treats of the structure and order of crystal- 

 lisation of the mineral components of these rocks. In 

 this chapter he embodies the substance of his paper on 

 the granular and porphyritic structure of massive rocks, 

 published some time since in the Neucs Jahrbuch. In 

 that paper he showed how in an eruptive silicate-magma 

 the minerals separate in the order of decreasing basicity, so 

 that at any given moment the uncrystallised magma is 

 more acid than the sum of the separated compounds. 

 Further, that the relative masses of the compounds 

 present in such a magma act only in so far on the order 

 of their separation, that generally those .'present in 

 less quantity crystallise out first. To facilitate petro- 

 graphical expression. Prof. Rosenbusch has proposed 

 a couple of words which appear to be worthy of general 

 acceptation. He calls those mineral components which 

 occur in individuals, bounded on all sides by crystallised 

 faces, idiomorphic ; allotriomorphic, those which owe 

 their boundaries to causes other than an internal 

 arrangement of the molecules composing them. Apply- 

 ing this to the plutonic rocks, he points out that, whereas 

 certain rocks, occurring as dykes, possess a " pan- 

 idiomorphic granular" structure, the plutonic rocks are 

 characterised by a granular structure essentially "hyp- 

 idiomorphic" (a part of the minerals only possessing their 

 own crystallographic form). 



The group of the Plutonic rocks is subdivided by Prof. 

 Rosenbusch as follows : — 



{(i) Family of the granitic rocks. 

 (*) Family of the syenitic rocks. 

 {c) Family of the eteolite syenites, 

 (rf) Family of the diorites. 

 ((?) Family of the gabbros and norites. 

 (/) Family of the diabases. 

 (yg) Family of the theralites. 

 [It) Family of the peridotites. 



Among these we notice a new name— the tlicralites. 

 Under this head are included the plagioclase-nepheline 

 rocks, formerly represented by the teschenites. The 

 latter ha\e been shown b)' Rohrbach (Tschermak's Min 

 und Pet. Mitt., 1885, ii. 1-63) to contain no nepheline, 

 and have consequently been referred by him partly to 

 the diorites, partly to the diabases. Still, plutonic rocks 

 representing this mineralogical combination appear to 

 exist (c'.^. in the Crazy Mountains in Montana, U.S.); 

 and thus the gap left by the removal of the teschenites 

 is filled up. Prof. Rosenbusch derives the name theralitc 

 from 6rn>iw (to seek eagerly). 



The group that embraces the rocks occurring in the 

 form of dykes, is subdivided, according to mineralogical 



