SCD S TOK. SIRO EANALS', SI 
direction of its transportation, are altogether conclusive against 
any such conception of its origin. Shore-ice could not account for 
the wide transportation of the drift, since shore-ice does little 
more than shove about, through short distances, such materials 
as are already at hand. It does not transport them in quantity, 
widely. The topographic distribution of the drift, too, seems to 
put insuperable difficulties in the way of the pan-ice theory. It 
does not follow that standing water and shore-ice have played 
no part in the production of the drift because they cannot be 
looked upon as the principal agents concerned. 
Icebergs. Icebergs emanate from glaciers. They may float 
out upon any lake or sea to which glaciers descend, and they 
may distribute themselves as widely as the conditions of tempera- 
ture, wind, and water movements allow. They may carry such 
stony materials as the glaciers from which they originated were 
possessed of. This they may distribute as widely as they are 
drifted. We have now to enquire whether icebergs, conjointly 
with the water of lakes and seas, could produce the drift of the 
North American and European continents. 
Icebergs cannot float beyond the limit of the lakes and seas 
on which their movements are dependent. The drift which they 
distribute cannot extend beyond these limits, and may fail by 
varying amounts, to reach them. These limits are vertical as 
well as horizontal. Berg drift can cover no lands which were 
not covered by the seas or lakes into which glaciers calved the 
bergs. The distribution of the existing drift by icebergs presents 
essentially the same difficulties, from the topographic stand- 
point, as its distribution by the sea. From the areal standpoint, 
the difficulties seem at first less overwhelming, since the southern 
and western limits of the drift might be thought to correspond 
with the geographic limits to which icebergs floated. On this 
hypothesis, the outer limit of the drift should be ill-defined, 
since very few bergs would reach the limit attained by those 
which went farthest. The distribution of icebergs today makes 
this point clear. If the drift were deposited by bergs, it should 
become thinner and thinner with increasing distance from its 
