
STRUCTURE OF ERUPTIVE ROCKS 189 
site been broken up and absorbed, one might have expected to meet 
with occasional detached xenoliths throughout the whole mass, while the 
granite itself ought to have varied more or less markedly in composition, 
‘considering the very different kinds of rock-material which it must have 
absorbed. Granitic batholiths do, indeed, sometimes vary remarkably as 
regards their petrographical character: but such variations are cases of 
magmatic differentiation, and appear to be due to the way in which the 
mineral constituents separated out from the original magma—so that in 
some places, particularly towards the margin of a mass, the rock is often 
more basic than towards the centre. It cannot be said, therefore, that 
the hypothesis of absorption is in all cases a satisfactory explanation of 
the phenomena. Quite recently, however, Dr Sederholm has shown that 
in Finland a process of fusion and assimilation has actually taken place. 
Certain large areas of schistose rocks—most of them of acid composition, 
but some markedly basic—have been melted and transformed into granite 
and granite-gneiss, scattered through which more or less numerous 
fragments of the basic rocks are conspicuous. (See oséea, p. 222.) Other 
geologists have speculated on the possibility of rock-masses having been 
pushed up or even blown out in fragments by vapours escaping from 
a batholith. But this would imply that all batholiths must have 
had communication with the exterior, which can hardly be admitted. 
On the contrary, there is no reason to doubt that many extensive masses 
of granite never had any communication with the surface, but cooled and 
consolidated at abyssal depths. On the other hand, the peculiar manner 
in which granite and other granitoid rocks are sometimes associated 
with effusive rocks, leads to the well-grounded belief that such batholiths 
are the cores or roots of ancient volcanoes. As an example may be cited 
the augite granite of the Cheviot Hills, which is closely associated with a 
great series of lavaform rocks and tuffs. In such cases it must be 
admitted that some batholiths are of less deep-seated origin than others. 
It must not be supposed that granite always occurs in 
boss-like masses. On the contrary, it frequently appears in 
the form of extensive sheets of very variable thickness—and 
it may well be doubted whether many of the plutonic masses 
which have been supposed to occupy more or less vertical 
funnel-like cavities are really of this character. So far as one 
can tell from what is exposed at the surface, the so-called 
“bosses” may simply be partially exposed sheets, some of 
which, however, must have a thickness of several thousand 
feet. A sheet-like structure is suggested by the fact that, far 
removed from the margin of a granite area, inliers of the 
same rock are not infrequently revealed in the beds of streams 
which have cut their way down through a great thickness 
of the metamorphosed rocks surrounding the central mass. 

