PETROGRAPHIC CLASSIFICATION 285 



of rocks, and their grouping for purposes of classification. The Committee 

 finds that there is Httle prospect of any one of the several existing schemes 

 finding general acceptance. Replies to the questionnaire indicate an 

 almost unanimous opinion that classification of igneous rocks must be based 

 on ascertainable facts — composition, texture, and mode of occurrence — 

 and that theories of origin must be excluded. Many petrologists are con- 

 vinced that no one scheme of classification is adequate to meet all require- 

 ments and advocate two (or more) classifications : one based on mineral 

 content, the other on chemical composition, i.e. a classification of magmas 

 rather than of rocks. Apart from these, the general opinion strongly 

 favours classification based upon mineral composition and texture.^ There 

 is even more general agreement that the naming of individual rocks must 

 be so determined. 



The Committee is of the opinion that rock nomenclature should be 

 independent of age, geographical distribution, and the nature of the asso- 

 ciated rocks. Exception should be made of certain dyke rocks, notably the 

 aplites and pegmatites. 



On the question of whether there should be three main divisions or only 

 two, opinion is almost equally divided. Similarly there is a sharp division 

 of opinion as to whether texture or mode of occurrence should be used to 

 separate the groups. In general, opinion favours either two divisions, based 

 on texture ; or three, based on mode of occurrence, i.e. habit. The 

 Committee suggests a compromise and advocates three main divisions 

 defined in terms of texture. 



When the three divisions are based rigidly on field occurrence it is 

 inevitable that two rocks, identical in composition and in all their physical 

 characters, should bear different names ; and that, on the other hand, two 

 rocks differing widely in texture and appearance should bear the same 

 name. Admittedly, field relations are frequently so obscure as to render 

 their interpretation a matter of personal opinion. The Committee thinks it 

 desirable that the personal factor should be eliminated as far as possible in 

 classification and nomenclature, and believes that the ambiguities referred 

 to above would disappear were nomenclature made independent of mode of 

 occurrence. Several of those petrologists who, in general, favour the other 

 course are apparently willing to go half-way, as they raise no objection to 

 the use of such terms as ' dyke-basalt.' 



The Committee finds that many petrologists, particularly among the 

 teachers of the subject, experience difficulty in defining and using the terms 

 ' plutonic,' ' hypabyssal ' and ' volcanic ' (or ' extrusive ') when used with 

 reference to individual rocks. This is notably the case with ' hypabyssal ' 

 — a misnomer, in that many dyke rocks have consolidated under the same 

 depth-pressure conditions as the plutonic rocks with which they are asso- 

 ciated. Further, both granite and gabbro (plutonic rock types) occur as 

 dykes (hypabyssal rock bodies) ; while many composite intrusions consist 

 in part of ' plutonic ' gabbro and in part of ' hypabyssal ' granophyre in 

 intimate association. 



The conception of the existence of three phases of igneous activity, 

 commonly referred to as plutonic (major intrusive), dyke phase and volcanic, 

 is of great value in mapping and in the interpretation of geological maps ; 

 but the Committee points out that duplication of rock names and other 

 ambiguities arise when these terms are applied to the roughly corresponding, 

 though actually not congruent, main rock groups. 



1 By ' texture ' the Committee means ' those features which depend upon the 

 size and shape, and arrangement and distribution of the component minerals.' 

 A. Johannsen, A Descriptive Petrography of the Igneous Rocks (1931). vol. i, p. 32. 



