300 
Subsidence of roof blocks probably was the chief cause of 
intrusion of the Ilimausak batholite. — The premises from 
which this conclusion is drawn may be summarized thus: — 
(1.) A strong evidence in favour of the hypothesis of 
intrusion by subsidence of roof blocks is shown by some of 
the general features of this batholite, viz. (a) the flat roof; 
(b) the more or less steep flanks abruptly cutting off the 
strata of the sandstone and the volcanic series from sea level 
to an altitude of 1100 meters; (c) the numerous reentrant 
angles and minor irregularities of walls and roof. 
(2.) The direct evidence, mentioned above, of subsidence of 
roof blocks. 
(3.) The results of the preceding discussion which shows 
that the other hypotheses are unable to constitute a satis- 
factory explanation. 
Since a relatively low specific gravity of the magma is the 
condition of the subsidence of roof blocks, it might be objected 
to the applicability of the hypothesis in the present instance 
that some of the abyssal rocks of Ilimausak are heavy. Thus 
certain nepheline-syenites rich in iron attain a specific gravity 
of 3:1. These rocks, however, owe their large quantity of iron 
to differentiation processes operating after the intrusion of the 
magma, and the low specific gravity of the naujaite offers a full 
compensation. The mean specific gravity of the batholitic rocks, 
calculated from the figures given in the tables of analyses, is 
lower than that of the augite-syenite which envelops the sunken 
sandstone blocks mentioned above. Thus no difficulties to the 
hypothesis arise from the presence of subordinate masses of 
heavy rocks. 
As Daty has pointed out the hypothesis of overhead stop- 
ing is not opposed to the assimilation hypothesis. The roof 
blocks which sink will gradually be subjected to higher temp- 
eratures, and it must be supposed, that at least some of them 
will melt, while marginal assimilation may be an important 
