290 
to Dary it is extremely probable that all magmas, with excep- 
tion of some of the basic ones, have a lower specific gravity 
than common rocks at the same temperature, and a sinking 
down of detached blocks is for that reason possible To ac- 
count for the breaking free of the roof blocks, it is assumed 
that local heating of the roof rock by contact with the magma 
will produce local expansion, resulting in the formation of cracks 
and fissures into which the magma makes its way and thus 
progressively detaches one block after another. 
A very striking instance of a batholite, the intrusion of 
which seems only to be intelligible on thé basis of this hypo- 
thesis, is the Marysville batholite in Montana as described by 
Barrett’, and in the opinion of the present writer the same 
may be said of the Ilimausak batholite?. 
Evidence of subsidence of roof blocks at Ilimausak. — A 
direct proof of the subsidence of roof blocks is found in the 
presence of numerous sandstone fragments, the largest of which 
exceeds 200000 cubic meters, in the augite-syenite at the foot of 
the Iviangusat Mountains, Kangerdluarsuk. This occurrence has 
been described in chapter III (p. 51), and it has been shown 
that the fragments can only be derived from above, and that 
they must have sunk down through more than 900 meters of 
magma. Another instance of roof blocks in the batholitic rock 
is observed at Nunasarnak (p. 75). 
1 J. BARRELL, Geology of the Marysville Mining District. Un. St. Geol. Surv., 
Professional Paper No. 57, 1907. 
The author on his first visit to the Ilimausak region, in 1900, arrived 
at the conclusion that the coming to place of this batholite was 
mainly due to subsidence of roof blocks, and in January 1902 expressed 
his view in a lecture in the Danish Geological Society. Shortly after 
this Dary published his paper on the mechanics of intrusion. and 
put forth the hypothesis in a much more elaborate form and on much 
broader grounds than the present author had intended, and the pub- 
lication of the view was therefore deferred until the present full 
report. 
