3,8 DISCOVERY REPORTS 



obtained for the mean number per 20 min. haul gives only a very approximate indication of the level 

 of abundance. This applies particularly to the high figure 300 miles north of the convergence in 

 summer which is based on only one observation, St. 1228, in December, at which a large catch of 

 post larval specimens were taken. 



The diagram shows the spread to the south as being greater in summer than during the rest of the 

 year. This is due to the inclusion of four lines of stations worked in the 90° E-i8o^ region where, as 

 will be shown below, the total range is considerably greater than in other sectors. 



Regional variation 

 In Fig. 5 mean numbers per 20 min. haul from a series of observations made along the Greenwich 

 meridian are compared with those from a similar series made between 90° E and 170° E. The data 

 have been treated in the same manner as in Fig. 4. 



Q J X j X . X y — 



-2S 2LXX 



so^E-iyo^E 



aOO TOO 600 100 400 

 HOUTM 



300 200 too .[f- 100 iOO JOG ■! 



'Ale 

 NAUTICAL MILES 



00 SOO boo 700 800 

 SOUTH 



Fig. 5. The distribution of adolescent and adult E. triacantha in relation to the Antarctic convergence in the Greenwich 

 meridian compared with that in the sector 90° £-170° E (see Fig. 4 for explanation). 



It will be seen that the total north-south range is considerably greater between 90° E and 170° E 

 than on the 0° line; it extends over 1350 miles in the former and only 700 miles in the latter. The 

 range over which E. triacantha is abundant is, however, similar in both regions, though in the Green- 

 wich meridian the northern and southern limits are some 150-200 miles further north. 



There is little doubt that the southward spread of E. triacantha in the South Atlantic (0°) is re- 

 stricted, at least to some extent, by the Weddell drift, the core of which crosses the Greenwich meridian 

 in about 60° S (Deacon, 1937, p. 28). The northern limit of the Weddell drift is not clearly defined, 

 but Deacon (p. 25) records two occasions in the mid-South Atlantic when a marked temperature 

 change has indicated a convergence between Antarctic water coming from the north of Graham Land 

 through Drake Passage and that from the Weddell Sea. One occurred in approximately 55° S, 22° W 

 and the other, less clearly marked, between 51° 20' S, 07° E and 52° S, 06° E. The boundary, being 

 further south to the west of these last two positions, is likely, in 0°, to occur between 53° S and 54° S. 

 This corresponds to a position some 130-250 miles south of the mean position of the Antarctic 

 convergence in 0° and it will be seen from Fig. 5 that the number of E. triacantha fall off considerably 

 in this region. Ommaney (1936, p. 297) found a similar distribution in Rhincalanus gigas Brady, 1883, 

 that is a falling off in numbers to a level of comparative scarcity in the Weddell Sea water. 



In the Indian Ocean (Fig. 5, 90° E-i7o° E), although the numbers are small beyond 400 miles 

 south of the Antarctic convergence, the total range extends as far south as the boundary between the 

 East and West Wind drifts. 



