AKXIVERSARY ADDRESS OF THE PRESIDENT. 67 



Our uomcnclature, then, after the recognitiou of these essential 

 distinctions, must further acknowledge the more accidental, viz., 

 the physical condition of the rock, whether hyaline or not, the re- 

 lation of its constituents, and the like. Many of these secondary 

 distinctions I should prefer to indicate by adjectival affixes, reserving 

 differences of names for marked differences in chemical constitution, 

 and, after that, of crystalline condition. The latter, I grant, refers 

 rather to a difference of environment than to a difference in essence ; 

 a holocrystalline, a hemicrystalliue, and a hyaline rock of the same 

 chemical composition have quite as near a relation one to another 

 as the larva, pu]3a, and imago of an insect ; but the marked 

 difference in aspect and structure justify us, I think, in a nominal 

 separation. The presence of an adventitious mineral not materially 

 affecting the chemical composition, peculiarities of structure or the 

 Uke, should, I think, be indicated by epithets. 



The number of minerals, as is well known, which enter into the 

 composition of igneous rocks frequently enough to entitle them to be 

 called " rock-forming," is but small ; for convenience they may be 

 grouped as follows*. 



I. Oxides of Iron, S)~c. 



Magnetite, haematite, ilmenite, chromite (with spinel). 



II. Magnesia-iron Silicates. 

 Olivine, enstatite (with hypersthene), augite and hornblende, 

 biotite. 



III. Alumina-alkaline Silicates. 

 Felspars (with nepheline and leucite), muscovite. 



lY. Free Silica. 

 Quartz. 



From this list such minerals as apatite, zircon, titanite, sodalite. 

 nosean, are omitted, though occasionally some of them may be re- 

 garded as rock-constituents, because they do not appear to have any 

 very important classificatory significance, and some are generally 

 associated with certain of the above named. Garnet also is ex- 

 cluded, though occasionally an important rock-constituent, because, 

 as it seems to me, we are not yet in a position to deal with it. If 

 an original constituent, and not an accidental one, due in some way 

 or other to contact-effects, I am disposed to regard it as a member 

 of No. III. 



The igneous rocks, then, as it appears to me, fall naturally into 

 the following grouping, commencing with the most basic and using 

 the existing nomenclature as far as possible. 



The Peridotite group consists of olivine with some members of I. 

 and commonly some representatives of II. Its simplest holocrys- 

 talline representative is dunite, essentially olivine and chromite. 

 (I should apply the name to the rock whatever representative of 

 Group I. were present.) The next marked variety is given by the 

 * I follow very nearly the classification proposed by Eosenbusch. 



