410 T. L. Walker^- Causes of Variation in the 



Art. XLII. — Causes of Variation in the Composition of 

 Igneous Rocks ; by T. L. Walker. 



[By permission of the Director of the Geological Survey of India.] 



Many attempts have been made to explain the causes of the 

 variation observed in the composition of igneous rocks. This 

 is particularly the case in " stocks " and dikes in which varia- 

 tion reaches its maximum. Frequently the border facies are 

 more basic, though this does not always hold true. The Car- 

 rock Fell massive in the Lake district in England, the Meissen 

 syenite eruptive in Saxony and some of the large nickel-bear- 

 ing stocks in the Sudbury Nickel district in Canada are good 

 examples of such differentiated eruptives. In all the above 

 cases the border facies are basic — gabbro, norite, diorite, etc., 

 while toward the center there is an increased acidity till syenite 

 or granite is reached. The transition can generally be traced 

 step by step, proving that the rocks so different chemically and 

 mineralogically belong to the same eruptive and form a geo- 

 logical unit. A. C. Lawson has shown that in the dikes of 

 the Rainy Lake district in Ontario the dike rock near the wall 

 is fine-grained and quite basic — toward the center the rock 

 becomes gradually coarser in texture and more acid in compo- 

 sition. Dikes do not show the very wide variation noted in 

 stocks. This may be due to their much smaller dimensions 

 and consequent more rapid solidification. 



I propose to review briefly the more common theories 

 advanced to explain the phenomena above referred to and to 

 call attention to the part which gravitation seems to play in 

 causing heterogeneity in eruptive rocks. 



During the last few years many theories have been advanced 

 to explain the differentiation of igneous rocks. It is probable 

 that most of these theories are applicable in certain cases, but 

 it is equally probable that no one theory gives a satisfactory 

 explanation of all the phenomena observed. Better results 

 may be attained by considering the manifold relations of fused 

 magmas and tracing the influence of each of these relations on 

 the homogeneity of the fused mass. The differentiated erup- 

 tive massive is the volume in which the history of these 

 changes is more or less imperfectly recorded. 



Fragments of country rock are often torn loose by the 

 ascending molten magma and absorbed, thus causing the margin 

 of the mass to be more acid or more basic than the more cen- 

 tral portions of the stock or dike. This is a cause of variation, 

 but cannot properly be spoken of as differentiation. It is not 

 to be regarded as a chief cause of variation. If the difference 

 in chemical and mineralogical composition were due to this 



