250 THE ANTARCTIC. 



water in a liquid state is entirely absent. In South 

 Georgia water-brooks are common in summer, and near 

 the German station a permanent spring even has been 

 discovered. Similarly there occur in summer on the 

 South Shetland Isles, which alone are somewhat better 

 known to us, brooks formed by the melting of snow and 

 ice. 



The non-melting of the snow is of necessity accom- 

 panied by a change in its transformation. If it does 

 melt near the coast, it is sure to do so somewhat more 

 vigorously at some elevation, leading on the islands, just 

 spoken of, to the formation of the firn-snow with which 

 we are familiar, albeit that the process will be more 

 sluggish owing to the reduced summer temperature and 

 to the much shorter periods. But on the larger tracts of 

 land farther south these circumstances are completely 

 altered. There the dry powdery snow is, even in the 

 height of summer and at no great elevation, driven about 

 by the wind and piled up in huge snow-drifts, and the 

 melting of the snow can be of no great moment. It is 

 certainly very doubtful if active melting of the snow, 

 such as leads to the formation of brooks, and as we 

 witness on the Greenland coast, can take place ; the 

 hollowed out forms of the inland ice observed by Dumont 

 d'Urville on Adelie Land may possibly be mere clefts 

 in the ice ; nevertheless the big icicles of the great wall 

 of ice prove that even in high southern latitudes vigorous 

 melting does take place. Until minute observations, 

 similar to those made in the north by Peary, Von 

 Drygalski, and others, have been made on the spot, we 

 can rely in our speculations on the formation of Antarctic 

 land ice solely on what has been actually seen, viz., the 

 icebergs, which are not inland ice. It is no longer 

 subject to doubt, that their origin is to be traced to the 

 ice covering of Antarctic lands terminating in the abrupt 

 perpendicular walls along the coast, spoken of in the 



