THE NATURAL HISTORY OF IGNEOUS ROCKS. 31 



most cursory view of the distribution of igneous rocks is 

 the frequent association of widely different types in the 

 same district or even at the same volcanic centre. Thus, 

 among lavas, we often find rhyolites and basalts in intimate 

 association ; among dyke-rocks, aplite and minette, or bos- 

 tonite and camptonite ; among intrusions of more plutonic 

 habit, granite or granophyre and gabbro. Such cases of 

 what may be termed paragenesis are of too frequent recur- 

 rence to be regarded as fortuitous ; the members of a pair 

 must be "complementary" rocks, connected with one another 

 by the ties of common origin. 



Several writers have endeavoured to trace some law of 

 general application in the succession in time of different 

 types of igneous rocks in a district or province. Besides 

 its obvious bearing' upon genetic considerations, any such 

 established order will tend to throw light upon many points 

 touched upon in the foregoing paragraphs, such, for instance, 

 as the association of particular rock-types with special habits 

 of intrusion or with certain types of volcanic eruption. That 

 the actual facts are far from simple is evident from the very 

 different empirical results arrived at by different geologists. 

 Thus Bertrand, from an examination of the history of 

 igneous activity in the European area, lays down a general 

 law of increasing basicity in the later products of each great 

 system of eruptions ; while Geikie, dealing more especially 

 with the British Isles, finds in general evidence of increasing 

 acidity. 



Having regard to the volcanic rocks only, a succession 

 which holds good in numerous districts, and has often been 

 pointed out, is what may be described as an order of in- 

 creasing divergence from an initial type. For instance, 

 the earliest lavas or fragmental outbursts being of inter- 

 mediate character, say, pyroxene-andesites, succeeding 

 eruptions have been on the one hand increasingly acid, on 

 the other increasingly basic, terminating perhaps with 

 rhyolites and olivine-basalts. The two lines of variation 

 have been in a general sense concurrent, and the order of 

 precedence between corresponding terms of the two diverg- 

 ing series is not always the same. This succession was 



