X PKEFACK 



we know of its work when it is at its lowest level. Frequently a river alternately 

 erodes or aggrades its course according as to whether it is in flood or at lowest level ; 

 so that if we saw such a river at its lowest level, or at any rate during the ebbing 

 of the flood, and noted that then it had no appreciable erosive force, we should 

 not be justified in concluding that it had not accomplished erosion in the past. And 

 as with water so it is with ice, with the exception of the phenomenon explained as 

 above by De Geer. The Antarctic ice is certainly not now in flood, and its power to 

 sculpture land is enormously diminished as compared with what it was when the 

 outlet glaciers were about 2500 feet thicker than now, the Ross Barrier 800 feet 

 thicker and 200 miles longer, and much of the island ice perhaps 1000 feet thicker 

 than at present. Nevertheless there is much to learn from present Antarctic ice 

 conditions which helps towards the interpretation of the phenomena of a past glacia- 

 tion in Pleistocene times in both hemispheres. For example, the Ross Barrier may be 

 compared with the North Sea ice sheet ; the glaciation of Gerlache Channel and Brans- 

 field Strait with that of the Irish Sea and of the Isle of Man ; the upthrust marine muds 

 on the flanks of Erebus with some of the British marine deposits pushed up from the 

 bottom of the Irish Sea, &c. Some details of this will be given in the summary at 

 the end of the second volume. No less entrancing than the glacial problems of 

 Antarctica is the enigma of its lost Andes. In West Antarctica the Andean rocks, 

 both sedimentary and eruptive, are typically developed and as typically folded, 

 attaining heights of 6000 to 8000 feet. In East Antarctica, with the exception, 

 perhaps, of the grano-diorites of King Edward VII. Land, Andean rocks and Andean 

 structures are absent, but there emerges from near the Soixth Pole a mighty block- 

 faulted range from 8000 to 15,000 feet high, with ei'uptive rocks typically alkaline, 

 and upper Palaeozoic coal measures, not folded, resting partly on Devonian, jjartly on 

 Cambrian, or in places on Pre-Cambrian rocks. Have the Andean faults swerved 

 away from the Andean fold lines and the Andean petrological belt, so as to yield a 

 block-faulted range meeting in a sort of tectonic virgation the true Andean fold lines 

 of West Antarctica ; or is the range of the Antarctic Horst entirely distinct from the 

 Andes, and does it trend to Prince Regent Luitpold Land and Coats Land, on the 

 eastern side of the Weddell Sea ? If the latter view is correct, a sea channel, as 

 suggested by Penck, may divide Antarctica into two portions, leaving Graham Land 

 with all West Antarctica and Carmen Land and King Edward Land in the condition 

 of a festooned ai'chipelago. A physiographic feature of great interest in the part of 

 Antarctica to which these notes relate is the development of an "ice divide" to the 

 west of the Antarctic Horst of Ross Sea. Though considerably lower than the ranges 

 of the Horst, the " ice divide" sends mighty glaciers through low gaps in the Horst 

 eastwards to Ross Sea. Is this " ice divide " at all analogous to the Pleistocene " ice 

 divide" of the Baltic Sea at a time when it sent ice westwards across the summit of 

 Areskutan, 5000 feet above the level of the Baltic, across the main range of Scandinavia 

 into the North Sea ? Greenland has obviously such an ice divide at the present day. 



