HYPOTHESIS OF CAUSE OF GLACIAL PERIODS 769 
On the eastern side of the pole of the winds, the earth move- 
ment tends to drag the bottom air away from the center and this 
is perhaps one of the reasons why the moving cyclones on that 
side show a tendency to dispersal, both in course and in force, 
as previously noted. 
Possible relation to terrestrial magnetism.— It is scarcely appro- 
priate to this paper, if it were in my power, to discuss the rela- 
tions of this oblique system of atmospheric circulation to the 
oblique system of terrestrial magnetism. Crudely stated, the 
notion entertained is that an atmosphere charged with electricity, 
circulating obliquely about a rotating earth which is inset with 
magnetic and magnetizable matter, might give rise to a mag- 
netic system which should express the dynamic resultant of the 
atmospheric circulation. A comparison of the magnetic and 
atmospheric charts shows so many points of resemblance, some 
of which are striking peculiarities, as perhaps to justify this ten- 
tative notion until the mystery of the earth’s magnetism be solved. 
Connected with terrestrial magnetism are auroral manifestations 
whose distribution is notably similar to that of the chief Pleisto- 
cene glaciation, as was remarked many years ago when first the 
progress of exploration outlined the extent of the glacial depos- 
its. It can hardly be presumed that either terrestrial magnetism 
or auroral displays have in themselves any causal connection 
with glaciation, but if they are dependencies of atmospheric cir- 
culation they become eminently serviceable to glacial students 
by affording a tangible concrete expression of the dynamic 
correlation of the atmospheric circulation, free from the intricate 
complexities of the latter; in short, a natural resultant at easy 
command. The verity of the notion must, of course, depend 
wholly on the outcome of magnetic investigations on their own 
lines, which happily are now being vigorously prosecuted. 
SUGGESTIONS RELATIVE TO MINOR GLACIAL OSCILLATIONS 
It has already been noted that besides the oscillations of 
epochal order there were subsidiary ones which left their record 
in a series of concentric moraines which corrugate the individual 
