Reviews—Prof. Zirkel’s Petrography. 567 
This is probably as good a classification as any that could be 
devised. Inasmuch as rocks do not fall into sharply-defined natural 
groups, it is impossible to invent a classification against which 
objections may not be urged. 
The subjects referred to above occupy three-fourths (635 pages) 
of the present volume. The remainder forms an introduction to the 
detailed study of the igneous rocks. Mode of occurrence is first 
discussed, and it is satisfactory to note that the true significance of 
the contrast in structure between the volcanic and plutonic rocks is 
clearly recognized. The attempt of Rosenbusch to raise the dyke- 
rocks to the position of a group, comparable in importance to the 
plutonic and volcanic groups, is unfavourably criticised. The minerals 
which enter into the composition of igneous rocks are next con- 
sidered, especially with reference to the laws of association. Micro- 
scopic research has led to important modifications in the old ideas as 
to paragenesis, but there are many laws relating to the association 
of minerals which possess a sufficient amount of generality to make 
them worthy of most careful consideration. 
The chemical relations of igneous rocks are discussed at length, 
and the views of Bunsen von Waltershausen, Durocher, Rosenbusch, 
H. O. Lang, Lagorio, Iddings, and others are fully described and 
critically examined. The phenomena attending the consolidation of 
igneous magmas are most carefully considered and the constituents 
are classified as follows, with regard to their order of development :— 
1. Minerals foreign to the eruptive magma. 
2. Minerals produced before the normal consolidation of the mass; 
é.g. quartz in quartz-basalts. 
3. Minerals due to the action of the magma on the constituents 
belonging to 1 and 2. 
4, Minerals formed as products of normal consolidation. 
5. Minerals formed during the last stages of consolidation, which, 
however, can scarcely be regarded as belonging to the ordinary 
processes of consolidation ; e.g. tridymite and fayalite in lithophysze 
in obsidian, ete. 
6. Minerals which do not belong to any of the above categories, 
and which cannot be referred to 7 with certainty. 
7. Minerals produced in the rocks after consolidation by secondary 
agencies, or deposited in cavities by infiltration. 
In dealing with the classification of the igneous rocks, the author 
lays down the general principle that mineralogical composition and 
structure are the two factors of primary importance. He thinks that 
geological age may still be employed in certain districts, and no 
doubt he is right; but whether it is desirable so to employ it in a 
general classification is a question on which difference of opinion still 
exists. The classification ultimately adopted is very similar to that 
of the first edition, except that geological age is relegated to a 
secondary place. The felspathic constituents are used as the means 
of defining the principal subdivisions, and three main groups are 
recognized. The first includes the rocks characterized by the pre- 
dominance of alkali-felspars, the second those characterized by lime- 
