A. J. Bull—A Hypothesis of Mountain Building. 365 
disturbed for long ages, normal faulting possibly indicating a state 
of slight tension. The phenomena of fissure eruptions indicates 
that parts of the crust have been in considerable tension, which 
is further borne out by the wide evidence Professor J. W. Gregory 
has gathered that fiords are tension clefts and mft valleys. It 
would be out of place to enlarge on this aspect of the matter, but 
I desire to indicate that geological evidence does not readily 
support the hypothesis of a contracting earth. 
The theory of isostatic adjustment may be regarded. as well 
authenticated from the work of Barrell in the United States and 
Oldham in India. The tendency to this adjustment must follow as 
a consequence of the simple principles of hydrostatics once it is 
conceded that there underlies the crust material that will flow under 
long-continued, but not necessarily great, stresses; on the other 
hand, the completeness of such an adjustment must be conditioned 
by the strength of the crust to resist distortion. 
It is evident from the many types of igneous rock that are met 
with on the surface that the interior of the earth cannot be 
homogeneous in composition, hence it may be assumed that radio- 
active elements are not eventy distributed, and that where they 
exist in greater quantity there will result local heating. Now, where 
this occurs between the crust and the Hayford compensation level, 
the result could be that portions of the crust might be elevated con- 
siderably in relation to the crust above the lesser heated regions. 
‘There is evidence that the change of density in the lower parts of 
the crust might be greater than that produced by mere thermal 
expansion, for instance, the density of one mode of a particular 
rock has been shown by Dr. L. L. Fermor to differ from that of its 
norm by as much as twenty per cent, so that the slow melting of 
such a rock might easily produce a considerable change of density. 
Such density differences might continue to increase during long 
periods of time during which material would be removed by 
denudation from the elevated tracts and deposited in the relatively 
depressed tracts, causing these to sink still lower, as has been the case 
in the Gangetic trough. These cofditions might be terminated by 
a convective movement, which would in a comparatively short time 
establish more stable conditions under the crust. A movement 
of this kind occurring in one place might well start movements in 
adjoining districts where unstable conditions had been attained, and 
so the convective movements might spread until they became 
world-wide. The effect of such movements would be to act as a drag 
on the under surface of the crust, drawing out the lower portions 
into schists and gneisses and disrupting the upper portion as it was 
moved. This would be most marked in the lower layers of the crust. 
These would be dragged into folds and puckered, while the higher 
layers would not be so much affected. The gaps thus formed in the 
lower strata would be filled by molten material from below, which 
might reach the surface in places. 
