WILLIAM B. DWIGHT. 117 
almost universally attribute the glacial phenomena of 
this country to glacier movements. 
These apparently opposing views are now likely to find 
their reconcilement in the probability that in the United 
States the phenomena were chiefly due to glaciers, while 
in Canada there is reason to believe that icebergs or 
ice-fields, floating on great bays or rivers, may have com- 
bined with the great glaciers, which were undoubtedly 
present, to produce the visible phenomena of glaciation. 
Thus it is supposed that a great gulf of the Arctic ocean 
may have extended southward a long distance, up the 
basin of the Mackenzie River, carrying icebergs which 
produced the glacial phenomena of a large part of that 
region. (See ‘‘Glaciers and Glacial Radiants,’’? by Dr. 
E. W. Claypole, in the American Geologist, February, 
1889, page 83.) 
2. The supposition that there was an enormously thick 
ice-cap covering the entire polar regions and extending 
south we!l into temperate latitudes, which is the basis of 
‘*Croll’s theory,’ is becoming pretty thoroughly dissi- 
pated by recent investigations. It seems now highly 
probable that there was not even a uniformly thick sheet 
of ice stretched across the northern part of the North 
American continent from Labrador to Alaska, producing 
the embossing and striations everywhere visible in the 
high latitudes of the United States and in Canada. For | 
it is now well understood that icebergs cannot be formed 
except from glaciers ; that glaciers cannot be formed in 
great and continuous ice masses without the combination 
of heavy precipitation of snow, with the right conditions 
of enduring cold, and also of a difference between the 
cold and moisture of winterand of summer. Heavy pre- 
cipitation of snow can occur only in regions of great 
moisture ; and where the maxima of moisture exist with 
the other conditions, there will be the maximum thick- 
ness of ice-sheets, and vice versa. 
67 
