786 T. C, CHAMBERLIN 
adjacent lands from the exterminating effects that might natur- 
ally be associated with a glaciation in the tropics. 
Relative to the more specific location of the Permo-Carbonif- 
rous glaciation, it may be noted that a tendency to form “lows” 
in India, Australia, and South Africa, not far from the ancient 
glaciated areas, is observable even under the present conditions. 
It is presumed that much more pronounced eddies were, in 
Permo-Carboniferous times, located on the borders of the cold 
belt formed by the intensified antarctic circulation and that these 
determined the areas of specific glaciation. 
Conditions in the northern hemisphere.—In the arctic regions, a 
very low temperature may be confidently postulated under the sup- 
posed conditions, because of the extent of the land and the absence 
of oceanic circulation between the high north and the equatorial 
regions. A Siberian climate of an intensified order is therefore 
assumed to have prevailed over the high latitudes of the northern 
hemisphere. This may doubtless have given rise to limited 
accumulations of snow in favored localities, developing into 
glaciers, but on account of the general low precipitation and 
the dryness of the atmosphere, giving rise to large evaporation, 
this may not have becomea pronounced fact. In so far as glacial 
deposits originated in the interior of the land they would be 
liable to destruction by surface denudation before submergence. 
The Permo-Triassic land period was long. Here and there, as 
already noted, there are phenomena which find their easiest 
explanation in glaciation and ice transportation. It is to be 
anticipated that, if this view be correct, further indications of 
severe temperature in northern latitudes will be developed in the 
course of future studies. It may be remarked that there is now 
nearly as much evidence of Permian glaciation in the northern 
hemisphere in regions away from mountains as there is of Pleisto- 
cene glaciation in the southern hemisphere in like situations. 
As remarked at the outset of this part of the paper, a really 
satisfactory discussion of the Permo-Carboniferous glaciation is 
impossible in the present state of knowledge. JI am by no means 
blind to the uncertain factors that inhere, at once, in the time 
