712 



SCIENCE 



[N. S. Vou XL. No. 1037 



which involve the character and origin of 

 igneous rocks and their antecedent magmas, 

 a method which is disclosed by the construc- 

 tion of the book, as well as by the statements 

 of the author regarding it. The work is 

 divided into three parts. The first broadly 

 treats of what the author considers the facts 

 which need explanation in a philosophy of 

 igneous rocks. The second contains a gen- 

 eral, " eclectic " theory of the subject. The 

 third outlines the results of applying this 

 theory to the so-called " facts " previously 

 mentioned. 



By way of introducing the " facts " in the 

 case the author devotes a chapter to the 

 Classification of Igneous Rocks, and thereby 

 reveals his lack of acquaintance with some of 

 the fundamental principles of modern petrol- 

 ogy, those based on the physical chemistry of 

 crystallizing solutions so far as known. The 

 chapter also demonstrates the inexactness, or 

 incoherence, of his logic, or his indifference 

 to the meaning of words, for on page 2 he 

 says that " Eeasons are stated for preferring 

 a classification founded on actual mineral 

 composition," and on page 9 it is shown that 

 it is not possible to determine the actual 

 mineral composition of igneous rocks, and that 

 recourse must be had to chemical analyses. 

 He then proceeds on the assumption that a 

 collection of rock analyses grouped by Eosen- 

 busch according to Rosenbusch's system of 

 mineralogical classification is a classification 

 according to the mode of a rock, the mode 

 having been defined as the actual mineral com- 

 position of a rock expressed quantitatively. 

 Rosenbusch himself states that his classifica- 

 tion is based on the most noticeable minerals, 

 in porphyritic rocks, with little or no regard 

 in some instances to the minute minerals in 

 the groundmass, which may form a large 

 part of the rock. The author's misuse of the 

 term mode as well as his statements regarding 

 the Quantitative System of Classification, 

 which he calls the Norm System, show plainly 

 his failure to comprehend the fundamental 

 principles both of this system of classification 

 and of the chemico-mineralogical relations in 

 igneous rocks on which the system was 



founded. This is further shown by his effort 

 to indicate its methods by a hypothetical 

 jumble of biological species. 



The author pays a high tribute to the leader- 

 ship of Rosenbusch in connection with rock 

 classification, whose system he professes to 

 adopt, with modifications of his own, but he 

 violates absolutely the essentials of the system 

 in the third part of his book, and he ignores 

 Rosenbusch's judgment on principles which 

 conflict with the main thesis of his " eclectic " 

 theory. Rosenbusch based his system on the 

 microscopical petrography of igneous rocks, 

 in which branch of petrology he was the ac- 

 knowledged leader, and it was not to be ex- 

 pected that in the later years of his brilliant 

 career he would have undertaken to recon- 

 struct his system of classification in the light 

 of new discoveries in allied branches of scienca 

 But it must not be supposed that he had no 

 appreciation of the march of events; toward 

 new or revolutionary ideas he held a conserva- 

 tive course, and upon one occasion in a dis- 

 cussion of new ideas with which he was in 

 sympathy he remarked to the present writer 

 that he must not introduce such changes into 

 his book suddenly, but gradually in successive 

 editions, for otherwise his readers would not 

 follow him. On the same occasion he volun- 

 teered the remark : " I do not know what the 

 future petrography will be, but it will be 

 quite difl^erent from what it is now," in 1890. 

 It was as though this great leader of a 

 wandering people had had a vision of a land 

 into which he himself was not permitted to 

 enter. 



For the rocks grouped together by Rosen- 

 busch Professor Daly calculates average anal- 

 yses, chiefiy from the tables of analyses pub- 

 lished by Osann, and assumes that these aver- 

 ages represent types of each group, the sub- 

 jective character of such calculations not being 

 considered by him as objectionable. A funda- 

 mental error in his procedure with respect to 

 igneous rocks appears in the misstatement, 

 copied from Rosenbusch, that coarse-grained 

 intrustive rocks differ from their correspond- 

 ing porphyry and lava forms by the relative 

 proportions of their chemical constituents. 



