382 ANTARCTIC VOYAGE OF THE BELGICA. 



islands are too high in proportion to the area occupied by the base, 

 and that therefore the mountains can not fail to pierce through the 

 coating of ice. The antarctic glaciers are not stationary, any more 

 than those of other regions, and though they remain perpetually 

 under the sway of winter, they still move on. The plasticity of the 

 ice prevents its accumulation beyond a certain limit of height, and the 

 mantles of ice must, even under extremely rigorous conditions of 

 weather, be limited in thickness, while all the forms of the antarctic 

 glaciers must be those of a semifluid mass. There are thus both ice 

 rivers and cascades, and also forms recalling the " corrie glaciers." 

 But all are alike buried beneath a mantle of perpetual snow, and bare 

 ice is nowhere seen. "Inland ice,' 1 properly speaking, does not exist 

 on the large islands of the Palmer Archipelago. On the other hand, 

 on Danco Land and Graham Land, it is only the mountains situated 

 near the coast which show themselves, while the whole interior of the 

 land lying eastward is completely buried under the inland ice. 



We must not, however, imagine that the antarctic lands are at the 

 present day as heavily loaded with glaciers as they might be, for traces 

 of a wider extension, dating doubtless from the Glacial epoch, are still 

 preserved. The presence of these vestiges of the Glacial epoch seems 

 to me remarkable for various reasons, and on this account I should like 

 to bring forward some facts in support of my assertion. Gaston 

 Islet, our eighth antarctic landing place, lying a mile from the coast, is 

 a huge roche moutonnee, perfectly polished on the surface. At the 

 time of our visit it was almost entirely bare of snow. Opposite this 

 islet, at Cape Reclus, there rises along the coast a large moraine 

 running from northeast to southwest. An examination of the map of 

 the lands discovered by the expedition shows that the direction of the 

 moraine is that of Belgica Strait, and we are led to the conclusion that 

 the glacier which produced this moraine must have occupied the strait 

 itself, which has at this point a breadth of 10 miles and a depth of 342 

 fathoms. Another argument is supplied by our seventeenth and 

 eighteenth landings. On Bob Islet, not far from Wiencke Island, we 

 discovered some well-preserved fragments of a moraine, from 15 to 20 

 feet high, resting against the sloping shore at a height of 80 feet above 

 the sea. This moraine has the same direction as the channel, and its 

 height decreases gradually toward the west. On it were some huge 

 blocks of gneiss, perfectly polished. The red granite is in the form 

 of rounded bowlders, and the same is the case with other rocks, while 

 the diorite is often angular. 



On the other side of Belgica Strait, exactly opposite the former 

 spot, we discovered a fine moraine on Banck Island. Its height was 

 65 feet, and its direction parallel to that of the strait. It rested against 

 the sloping side of the mountain, which here displayed characteristic 

 roches moutonnees. These moraines can only be explained as the prod- 



