﻿Vol. 66.] CLASSIFICATION OF IGNEOUS ROCKS. 503 



VIII. Summary. 



(1) A 'natural' system of classification is one giving expression 

 to facts of nature, the plan of the system being demanded by the 

 laws of the origin of the principal characters of the objects to be 

 classified. Natural laws are established by facts, not by process of 

 reasoning. 



(2) Xo current classification of igneous rocks is, in any of its main 

 features, in accord with the facts of nature. 



(3) It is questioned whether petrographic system for igneous 

 rocks can be strictly natural, because all their important characters 

 are gradational within a group, and are due to complex geological 

 conditions. 



(4) The scientific logical classification of igneous rocks must 

 apparently be based on the quantitative development of funda- 

 mental characters, and the divisions of the scheme must have 

 sharp artificial boundaries, since none exist in nature. 



(5) Chemical composition is the fundamental character of igneous 

 rocks, but it may be advantageously expressed for classificatory 

 purposes in terms of simple compounds, which represent either rock- 

 making minerals or molecules entering into isomorphous mixtures 

 in known minerals. It is probable that the magmatic solution 

 consists of such molecules and that the norm of the Quantitative 

 System is a fairly representative set of these compounds. 



(6) The actual mineral and textural characters of igneous rocks 

 are variable qualifiers of each chemical unit, and should be applied 

 as such to terms indicating magma tic characters. 



Discussion. 



Mr. Harker welcomed this paper as a clear statement of the 

 Author's position. He took a more hopeful view than the Author 

 of the ultimate prospects of a natural classification of igneous rocks, 

 and believed that in Vogt ? s conception of ' anchinioiiomineralic ' 

 and ; anchieutectic ' types we had already the germ of such a system. 

 He was content to await in patience the gradual development of a 

 more rational classification with increasing knowledge. The con- 

 viction that this development would be seriously impeded if a rigid 

 artificial taxonomy and terminology came to be generally adopted, 

 was one ground for his objection to the Quantitative Classification. 



The speaker thought that the authors of this system had been 

 decidedly, if unconsciously, influenced by the now obsolete concep- 

 tion of a rock-magma as a mixture of free oxides, which might be 

 arbitrarily linked together for purposes of convenience. He pointed 

 out that, since such constituents as lime and alumina enter into 

 ferromagnesian as well as into felspathic minerals, the choice of the 

 ' standard' molecules may affect even the class to which a given 

 rock is to be assigned. Thus the ' hornblendite ' of Brandberget, 



