290 DISCOVERY REPORTS 



THE ANTARCTIC ICE-EDGE 



The Antarctic pack-ice generally has a well-defined northern boundary. A ship 

 steaming southwards in the Southern Ocean will at most times and places pass through 

 open water until close pack-ice comes in sight. That is to say, there is not usually a 

 gradual transition from isolated streams of ice to close pack. Such conditions indeed 

 may sometimes be found, but far more often the ice is first seen as a solid white line on 

 the horizon which soon bars further progress to a ship not prepared to force a way 

 through it, and if the ice is first seen as an isolated stream it can generally be assumed 

 that the main body of the pack is not far off. 



Open water may of course lie to the south of the ice-edge. It is well known that open 

 water is expected to be found in summer beyond the belt of pack in the Ross Sea, and 

 in the Atlantic sector more or less open water is formed about December in com- 

 paratively high latitudes, while a belt of pack still lies to the north. In the Weddell Sea 

 conditions appear to vary very much from year to year, and the distribution of ice is not 

 predictable in the present state of our knowledge. Little is really known of the interior 

 of the ice belt, but it seems probable that as a general rule there is more ice than open 

 water between the ice-edge and the shores of the Antarctic continent. This paper, 

 however, is almost exclusively concerned with the position of the northern boundary of 

 the pack. 



The type of floes seen at the ice-edge varies with the time of year and to some extent 

 with the locality. In spring and summer small floes and brash are commonly met with, 

 though much larger floes are to be found a little distance within the ice beyond the action 

 of the swell. In autumn and winter the outer fringe of the pack often consists of ice 

 scum which merges into pancake ice which in turn gives way to more substantial floes. 

 The thickness of the ice varies considerably in different localities. Floes which have 

 drifted from the Weddell Sea tend to be hummocked, overridden and very heavy. In 

 parts of the Pacific sector, where it appears that the ice may become locked in the 

 position of growth for considerable periods, immensely thick floes are to be seen. 



The ice-edge, however clearly demarcated, lies in a tortuous line, often with deep 

 bays and with promontories which may extend some miles from the main body of the 

 pack. In the charts accompanying this paper most of the records of the positions of the 

 pack-ice are derived from isolated observations or a series of recorded positions. 

 Occasionally, however, a ship has followed the course of the ice-edge and charted it in 

 detail for a considerable distance. Thus the 'William Scoresby' in October 1928, 

 charted the outline of the pack east of South Georgia, and in Plate LXXIV, between 28 

 and 3 2° W, the winding course of the ice-edge is well exemplified. The ' Pourquoi Pas? ' 

 similarly followed the edge of the pack for a great distance in the Bellingshausen Sea, 

 and this record is plotted in Plate LXXXV between 76 and 123 W. 



There are various signs by which a ship steaming southwards may be warned of the 

 proximity of the pack before the ice itself is in sight. The Antarctic Petrel is nearly 

 always seen about 400 miles or less from the ice-edge. The appearance of the Snow 



