Herman G. Simmons 
during the whole tinie of its existeiice. The marks of glaciation preserved iip to 
our days poiiit to two different centres, one on each side of Hudson Bay. Now 
most probably these marks must be atlHbuted to the last stages of the iceage, but 
it seems very probable also that tlie inland ice, uniting from the two centres at 
the height of glaciation, as doubtless it did, must have in its gradual melting away 
passed very similar stages to those of its progress, as the highlands which held 
the last reinnants of the icesheet must certainly have doiie so because the accu- 
mulation of ice was greatest in the centres and had beeu so from the beginning 
of glaciation. 
During earlier stages of glaciation, before the eastern (Labrador) icesheet and 
tiie western (Keewatin) closed togetlier, a southward way of escape must have 
existed, where the plants from the Hudson Bay area and the eastern islands of 
the Archipelago could rctreat. A similar passage doubtless was to be found for a long 
time between the Keewatin inland ice and the Cordilleran icesheet; this may have 
been of a very great im))ortance for the circumpolar species in reaching southward 
aloug the whole Cordilleran range, but even if it has certainly played a part in 
the reimmigration also. it has perhaps not had the importance that might easily 
bo attributed to it in stocking the present arctic parts of the Continent and 
the islands with western american species, for, as is well known, there existed 
a wide area from the mouth of the Mackenzie to and beyond Bering ötrait which 
was Jiever glaciated, and where climatic conditions may have been such as to allow 
vegetation to live over the iceage. The existence of this large icefree area in Alaska 
has, indeed, puzzled me rather much, but 1 have at last found an explanation which 
I think may be the right one. 
If we look on Chamberlin's or Wrights maps of glaciation we shall find the 
Cordilleran icesheet ending in the M'Kinley range or south of it, which would seem 
(juite unacountable. l)ut if we go to the text or to papers dealing with the surface 
geology of the Alaskan Peninsula and the Aleutian Islands, facts will appear which 
throw liglit over the problem. Wright, for instance, in Iceage in N. Am.. p. 15, 
says (speakmg about Mount Shasta): »But from this point on, glaciers multiply and 
continue, in ever-increasing glory, through the Coast Range of British Columbia 
Alaska to the islands of the Aleutian Archipelago», and p. 30: 
»Elliott estimates tli! 
iitmg great and small, there can not be less than five 
thousand glaciers between Dixons Entrance and the extremity of the Alaskan Penin- 
sula». He alsospeaks of a glacier on Unalashka and maps one on Uni mak Island. 
But where in present time glaciers are to be found we must have the right of 
concludmg tliat a inore comprehensive glaciation has existed during the iceage, 
;n)<l Kl ssKLi,. (TtM»!. Alaska, sjieaks of many traces of it in Unalaskha and other islands. 
HeiL-, I think, we have the clue to the absence of glaciation in nortliern 
Alaska. At the time when the Aleutian Islands formed a continuous mountain 
Chain, this caught up the moisture of the winds coming from the Pacific, and it 
was deposiled here as nevés and glacier tongues, even if it never reached to form 
