320 SCIENCE PROGRESS 



leucite is associated with the intermediate plagioclase felspars 

 but not with albite. If there is not enough silica present to 

 form both orthoclase and albite with the alumina, potash and 

 soda, the potash as the stronger base will take the larger 

 amount of silica, so that orthoclase (K 2 . Al 2 O s . 6Si0 2 ) and 

 nepheline (Na 2 0. A1 2 3 . 2Si0 2 ) are formed. If, however, the 

 elements of anorthite are present its crystallisation induces the 

 simultaneous formation of its isomorph, albite (Na 2 O.Al 2 3 . 

 6Si0 2 ), and the two crystallise out in solid solution of one 

 another, forming a member of the plagioclase series. As there 

 is now insufficient silica to form orthoclase, leucite (K 2 . Al 2 O s . 

 4SiO) appears in its place. 1 A concise but adequate account is 

 given of the chemico-physical principles applicable to silicate 

 magmas and their consolidation to form rock masses. 



The chapter dealing with the crystallisation and texture of 

 igneous rocks, illustrated by excellent photographs, is one 

 of the most valuable in the book. It includes a description of 

 the different " habits " of the principal rock-forming minerals, 

 according to the conditions under which they were crystallised. 

 A number of novel terms are introduced, and some at least of 

 these might be adopted with advantage. 



After dealing with the mode of occurrence and larger struc- 

 tures of igneous rocks, special attention being given to the 

 mechanics of columnar jointing, the author devotes a large, 

 section of the work to the subject of the nomenclature and 

 classification of the igneous rocks. We are given in the first 

 place a " qualitative mineralogical " classification based on the 

 works of Rosenbusch and Zirkel, which may be found useful 

 for reference, though the essential constituents are not always 

 sufficiently distinguished. This is followed by a detailed descrip- 

 tion of the newly elaborated " quantitative classification " of 

 igneous rocks, already reviewed in Science Progress (vol. i. 1906, 

 p. 259). It is unnecessary to add anything here to the severe 

 judgment which Mr. Harker has passed upon this scheme : 

 " The rounded completeness, the measured precision, the finality, 

 which would be admirable in a mathematical treatise, serve only 

 to condemn a classification of igneous rocks, since they make 

 more evident its aloofness from the scheme of nature, based not 

 on arithmetical but on physical and chemical principles." 



1 This agrees very well with the supposition {ante, p. 308) that the alkalies are 

 usually present in magmas in simple combinations with alumina and not as felspars. 



