BELT ON THE GLACIAL PERIOD IN NORTH AMERICA. 99 



the heat of the Gulf Stream and other warm currents that is now 

 expended in tempering Arctic seasons and melting polar ice, Avould 

 then be spent in evaporation ; and greater evaporation would give 

 greater precipitation on the frozen lands of the north. The forma- 

 tion of continental ice requires both a low mean temperature and an 

 abundance of moisture. The rise of northern lands, and the closing 

 of the Arctic basin to southern waters, would give the one, and the 

 greater warmth of tropical and sub-tropical seas, into which no polar 

 currents ran and from which no warm waters flowed, would give 

 the other. 



IV.— ACTION OF THE ICE. 



1. Statement of the question. — Turning our attention now to the 

 possibility of land ice having glaciated the whole of eastern North 

 America, we encounter the difficulty that has prevented many from 

 accepting the theory of continental ice, who are fully impressed with 

 the satisfactory solution it affords of the distribution of the drift-beds 

 and the erosion of valleys. It is that whilst the rock scratchings 

 and transported blocks prove that the glaciating agent moved from 

 the north, there are no mountains in that direction from which it 

 could have descended, and that any elevation of x'irctic regions suf- 

 ficient to give a slope that would bring the ice southwards like a 

 great glacier is utterly improbable, if not physically impossible. 

 With this opinion I fully coincide, but so far from considering it 

 fatal to the theory of land ice, I believe that no such slope was 

 necessary, and that the theory better explains the phenomena of 

 the drift, on the supposition that there was no great elevation of 

 northern lands than by a contrary hypothesis. 



The subject may be best understood by tracing in imagination 

 the accumulation of the ice and its progress southwards, its cul- 

 mination and subsequent retreat, and noting whether or not its pro- 

 bable mode of action will account for the facts to be explained. It 

 will be convenient to limit the discussion to one great area such as 

 that of eastern ISTorth America, where the glaciation though on a 

 grander scale than elsewhere, is more uniform from the very vastness 

 of the agent that effected it. 



2. Accumulation of the ice. — As the glacial period, from whatever 

 cause, came on, snow and ice would gradually spread from the arc- 

 tic circle southwards. Wherever there was not inclination of the 



