ON A HILLSIDE IN DONEGAL 353 



Pressure-changes are here traceable, but the main features 

 are clearly original, and subsequent attempts at shearing have 

 obscured the gneissic structure rather than intensified it. This 

 is in accordance with the views of Bonney, already quoted. 



So far, perhaps, a case has been stated, and it is well to 

 return towards a more judicial attitude. We have from time 

 to time noted the inclusion of dark blocks in our typical gneiss 

 or granite, and have without question regarded them as 

 remnants of some pre-existing rock through which the granite 

 broke. But this is very far from representing the prevalent 

 opinion of geologists. Every crystal or group of crystals in 

 an igneous rock represents a separation of material, a segrega- 

 tion, whereby the surrounding magma is left the poorer by some 

 constituents. From various causes, mere gravitation being one 

 of them, at some time or other in the earth's history, and perhaps 

 in every caldron of molten rock beneath our feet, a differentiation 

 of material may have taken place, giving us a magma rich in 

 silica on the one hand, and a magma rich in ferro-magnesian 

 silicates and iron ores on the other. 1 Long before BrOgger so 

 brilliantly developed the differentiation-theory, in his great paper 

 on the " Ganggefolge des Laurdalits " (1898), to account for the 

 series of igneous rocks in the neighbourhood of Christiania, 

 " segregation-veins " and " segregation-patches " had been freely 

 spoken about, and it had been felt that mineral aggregates of 

 very varied nature could separate out from a once homogeneous 

 mass. J. A. Phillips, 2 in discussing the lumps of various kinds 

 found in granite, referred the more crystalline and granitoid 

 ones to segregation-patches, and the more angular and 

 obviously sedimentary blocks to inclusions picked up during 

 flow. Although the margin of a block might be very sharp, 

 yet a crystal from the granite might be found running into it ; 

 and this was rather naturally accepted as evidence that the two 

 types of rock consolidated from the same magma. We have 

 already seen that this is by no means a necessary conclusion. 

 The microscopic study of contact-products had, however, not 

 progressed very far even five-and-twenty years ago, and the 

 literature of true inclusions has been immensely enriched since 



1 An admirable discussion of this question, on which so much has been written, 

 is to be found in the fifth chapter, " Die Differentiation der Magmen," of Doelter's 

 work on Petrogenesis (1906), p. 71. 



3 Quart. Journ. Geol. Soc. London, 1880, p. 1, and 1882, p. 216. 



23 



