176 DISCOVERY REPORTS 



Work in the Weddell Sea revealed several distinctive associations closely correlated 

 with the history and movements of the surface waters. To the north-east work along the 

 ice-edge in early spring showed that the main increase had not begun in October. 

 Development in the eastern Weddell Sea water apparently reached its maximum about 

 mid-season, as Gran (1932, p. 352) has already placed on record. Early in January 1931 

 vast numbers of the three large species previously mentioned were encountered between 

 South Georgia and the South Sandwich Islands, and thence south-eastwards as far as 

 63° S. Rhizosolenia styUformis had the most southerly distribution of the three, Corethrofi 

 voldiviae predominated near the old pack-ice in about 57° S, and Chaetoceros criophiltim 

 was less regular in its distribution. South of lat. 63° a poorer phytoplankton with 

 Nitzschia seriata dominant was found up to the circle, and south of that again a still 

 poorer phytoplankton, somewhat similar to that found in the far south of the Bellings- 

 hausen Sea, was found close up against the fast ice within about 100 miles of the 

 Antarctic mainland. Here the dominant species were Chaetoceros socialis, Ch. neglectiis, 

 Nitzschia seriata and A'^. closterium. 



The water issuing from the north-western corner of the Weddell Sea, which probably 

 passes through the last stage described above about two years previously, and which 

 forms the eddy round Joinville Island into the Bransfield Strait, was found in December 

 to be supporting a phytoplankton precisely similar in quality to that found to the south- 

 west of South Georgia a month earlier, but by no means so rich. Later in the season 

 work farther to the westward showed that the phytoplankton in this type of surface 

 water had lost its distinctive character, merging imperceptibly with that of the eastern 

 Weddell Sea and Bransfield Strait. Towards the end of the season the phytoplankton 

 to the north-west of the Weddell Sea area became poor, while to the north-east it re- 

 mained rich, partly owing to the drift of the aftermath of the main increase farther west. 

 In the abnormally warm 1929-30 season a very poor phytoplankton prevailed over the 

 whole of the northern part of the Weddell Sea area in March. 



In Bransfield Strait an association of almost pure Corethron valdiviae was found 

 at all seasons in the old water of Bellingshausen Sea origin that occupies the greater part 

 of the strait. Other species were most frequent in this water in spring, when the associa- 

 tion is probably regenerated within the strait itself, but they never formed any notable 

 proportion of the total numbers. The phytoplankton of Bransfield Strait was poor in 

 comparison with that of other areas in the same latitude, and though by far the richest 

 hauls were obtained in November, signs are not wanting that in most years the main 

 increase begins somewhat later than at South Georgia, as one would expect from its 

 more southerly latitude. The phytoplankton tended to be richer at the eastern end of 

 the strait ; but the eddy of Weddell Sea water contained a very different phytoplankton 

 during the first half of the season, and this appeared to reach its maximum somewhat 

 later in the year than that of the Corethron plankton which occupied most of the strait. 

 An autumnal maximum due to invasion from the Bellingshausen Sea probably occurs 

 in most years. 



The material from the Bellingshausen Sea was obtained in higher latitudes than that 



