254 SCIENCE PROGRESS 



words, it is a factor which makes for a certain standard se- 

 quence of crystallisation of the several rock-forming minerals, 

 and it serves in part to account for such measure of truth as 

 can be claimed for the empirical laws laid down by Rosenbusch. 

 Limitations of space preclude any discussion of other 

 branches of our subject, and in particular of the application 

 to rock-magmas of Bakhuis Roozeboom's investigations on 

 isomorphous mixed crystals. Enough has perhaps been said 

 to show the interest which attaches to this view of the process 

 of crystallisation in igneous magmas, and the promise which 

 it offers of further development. When a sufficient body of 

 quantitative data has been accumulated by research in the 

 laboratory, the solution theory of rock-magmas will become the 

 proper physical basis of all investigation and speculation in 

 the domain of petrogenesis. 



