202 DISCOVERY REPORTS 



more conspicuous in these months than at other times of year. It seems probable that further examina- 

 tion of such features as these will show that they are correlated with the effects of bottom topography. 

 Between about io° W and 30° E the positions of the isotherms are very much influenced by the 

 eastward flow of cold water from the Weddell Sea, which also has an important effect here on the 

 distribution of pack-ice in the early summer (see Mackintosh and Herdman, 1940, p. 293). In winter 

 and spring the isotherms tend to be pressed up towards the convergence which here lies about its 

 lowest latitude, but in December, while pack-ice and cold water persist in a relatively low latitude 

 between o and 30° E, east of 30° both ice-edge and isotherms bend far to the south, and the ice belt 

 begins to break up internally, leaving an outer zone which contracts towards the South Sandwich 

 Islands (compare the ice-edge in Plates IV and V, December and January). As the outer zone of ice 



60 ■ 



S-6 - 



40- 



3-0- 



2-D- 



1-0- 



00- 



-l-O- 



Fig. 10. Comparison of average monthly temperatures, based on monthly isotherms. 



melts away it tends to leave a long tongue of cold water which is seen as a conspicuous S-shaped turn 

 in the isotherms in Plates V-VII. The shape of the isotherms here is subject to a good deal of variation. 

 The data indicate that a line of observations running south from the convergence will usually, but not 

 always, reveal a slight rise in temperature about 60-65° S, and if there is no actual rise in temperature, 

 there will at least be a long expanse of ocean in which the temperature at the surface will change very 

 little. The existence of a belt of relatively warm water here, which melts the ice in a high latitude while 

 a zone of pack-ice still persists to the north of it, is noted by Deacon (1937, pp. 18, 28, etc., and Fig. 8), 

 who describes it as an ill-defined divergence region with an upwelling of warm water between the 

 Weddell Sea current moving eastwards in a lower latitude, and the current moving westward in a 

 higher latitude near the continental coast. Deacon also refers (p. 30) to the outer and inner belts of 

 pack-ice, and adds : ' It is just possible that even in winter there may be open water between the two 

 ice-streams in the eastern part of the Atlantic Ocean.' 



