3o8 DISCOVERY REPORTS 



flowing westwards into the bight south of 66° S and known as the East Wind Drift current 

 and from the South Atlantic warm deep water. The main body of the population, as 

 sampled in the colder water carrying ice, is probably carried into the Weddell Sea in the 

 East Wind Drift, but where water from the Weddell Sea meets with warmer water from the 

 Bellingshausen Sea, north of the Scotia Arc and farther east as far south as St. 808, a sparse 

 population is found which owes its origin to Atlantic warm deep water. The population 

 in the East Wind Drift flowing into the Weddell Sea, in its turn, originates in warm deep 

 water in the South Indian Ocean. It appears to rise to the surface before leaving the 

 Weddell Sea in the north-easterly current and this, at least partly, may be due to the 

 upwelling of warm water which takes place in the centre of the circular Weddell Sea 

 current system and along the divergence region between the two water movements 

 which make up that system. 



These conclusions are largely in agreement with those of Ottestad (1932), who has 

 already suggested that the stock of Rhincalamis in these waters belongs not properly to 

 the Weddell Sea but to waters outside that area and is carried into it by southward- 

 moving water from the South Atlantic. 



HORIZONTAL DISTRIBUTION, SEASON 1932-3 

 Falkland Sector, October to December (Table III 6) 



The lines of equal distribution of R. gigas for the first half of the season 1932-3 

 (Figs. 14, i^) show certain differences from those for the season 193 1-2. The area of 

 maximum abundance is greatly reduced, and only at two stations (1017 and 1025), 

 between the Falklands and South Georgia, did the catch in the combined nets exceed 

 10,000 individuals. The line of stations at the western end of the Drake Passage was 

 taken during the last week in October, over a month earlier in the year than in the 

 previous season. As we have seen (pp. 301-3) the spring ascent to the surface from the 

 winter level of the species was not yet completed at that time. The comparatively small 

 numbers taken on this line may therefore be attributed to the fact that Rhincalanus had 

 not yet completed its ascent into the range of the 250-100 and loo-o m. nets when the 

 line was taken. 



In the spring of 1932-3 the Bransfield Strait was found to be clear of pack-ice, the 

 edge of which lay farther south in the Bellingshausen Sea than at the same time of year 

 in the previous season. The most southerly station — St. 735, on the Western Drake 

 Passage Hne in November 193 1 — lay in 63° 55' S and 73° 28-8' W (Fig. la), while at 

 the end of October 1932 the most southerly station, at the edge of the pack-ice, was 

 taken in 66° 457' S and 80° 19-8' W (St. 994, Fig. 2 a). In October 1932 and early 

 November the edge of the ice was followed into the Bransfield Strait to Deception Island 

 and finally as far as the longitude of Joinville Island (Fig. 2 a). R. gigas, and indeed the 

 Copepoda generally, were found to be practically absent from the Bransfield Strait and 

 from the coldest Bellingshausen Sea water (with a temperature lower than — 1-5° C, 

 carrying unmelted pack-ice). The plankton as a whole in this area was extremely poor. 



