28 



DISCOVERY REPORTS 



■ '• was frequent iy encountered. The appearance of the drift-ice and the rapid fall in 



ice temperature both suggest that the water south of 51-52 S, within 200 miles of 



:as a different origin from that farther north. Its properties, and the 



ice that it carries, suggest that it belongs to the Weddell Sea current, whilst the warmer 



water farther north contains the water from the Bellingshausen Sea and the Pacific 



Ocean. Another temperature record, made during a voyage towards the north in 



14-15 E, showed that north of 55 S the surface temperature increased gradually, and 



there was no indication of a convergence south of the Antarctic convergence. The 



boundary between the Weddell Sea and Bellingshausen Sea currents, as far as it can 



be decided from the existing data, is shown in Fig. 4 (p. 19). 



The coldest water in the Weddell Sea current is found 400-500 miles to the south of 



this northern boundary ; the figures in Table I show its position in the lines of sections 



5, 6, and 7. 



Table I. 



The cold water in the lower stratum of the surface layer at St. 850 in 50° 44' S, 

 31 44' E (section 8, Plates XIII-XV) seems also to belong to the Weddell Sea current, 

 but the surface temperature distribution, illustrated in Fig. 8, shows that as far east 

 as this the current has almost lost its identity; farther east it can no longer be 

 distinguished from the main body of the easterly drift. 



Longitudinal sections east of 30 W show that between the Weddell Sea current and 

 the well-defined westward movement near the continent, there is a tongue of warmer 

 water. It is shown very clearly by the surface temperature chart in Fig. 8. The sections 

 made across the region occupied by this warmer water show that it is an ill-defined 

 divergence region and an area of irregular movements between the two opposite currents ; 

 the existence of the warm tongue of water is, however, reliable evidence of a pre- 

 dominant movement towards the west. 



The movement towards the west, the northward current along the east coast of 

 Graham Land, and the current flowing out of the Weddell Sea towards the east, form 

 three parts of a cyclonic movement which extends across the entire width of the Atlantic 

 Ocean. The surface temperature distribution indicates that the cyclonic movement may 

 be completed by a southward movement between 20 and 40 E ; there is, however, very 

 little evidence of such a current at the surface ; the conditions are not very different 

 from those farther west and more in keeping with the existence of a small northward 

 movement. The chart given by Fricker (1898), reproduced in Fig. 9, shows that the 

 northernmost limit of pack-ice also bends towards the south in 30-40 E ; the same can 

 be seen to be true of the actual edge of the pack-ice in seasons when there are sufficient 

 observations for it to be plotted (see a chart by Daehli, 193 1). The curving of the pack-ice 



