RHINCALANUS GIGAS 313 



the catches. It is again evident that while the Drake Passage was an area where R. gigas was 

 the most important species of copepod in the fauna (75 per cent and over), the Weddell Sea 

 was an area where it made up only a small proportion of the fauna. At the stations in the 

 South Sandwich-South Georgia area with a temperature lower than 0° C. it comprised less 

 than 25 °/o of the total Copepoda and at all stations with a temperature less than — i -0° C. 

 less than 15 per cent of the total. It will be seen immediately that there is a striking 

 difference between the percentage distribution (Fig. 15) and the numerical distribution 

 (Fig. 14). In the season 193 1-2 the two figures for the numerical and the percentage dis- 

 tribution (Figs. 8, 9) resemble one another fairly closely. Throughout the whole of its area 

 of maximum abundance in that year (catches of more than 10,000 individuals) R. gigas 

 was the dominant species of copepod, forming 75-100 per cent of the total copepod 

 catch. In the season 1932-3 R. gigas was the dominant species of copepod in the Drake 

 Passage (Fig. 15), but farther east between the Falklands and South Georgia it formed 

 only 50-75 per cent of the catch at several stations where more than 5000 individuals 

 were taken in the two hauls together. At St. 1025, where more than 10,000 individuals 

 were taken, it formed less than 50 per cent of the catch. It is evident that in the northern 

 and eastern part of the area of abundance in the season 1932-3 species of copepod other 

 than R. gigas were present in larger proportions in the catches than at corresponding 

 stations in the preceding season. Thus at Sts. 1019, 1023, 1025 and 1029, between the 

 Falklands and South Georgia (Figs, za, 15), species other than R. gigas formed between 

 25 and 50 per cent of the copepod fauna, while at Sts. 750, 751 and 788, which are as 

 nearly as possible the corresponding stations in the previous season (Figs, i a, 9), species 

 other than R. gigas formed less than 25 per cent of the copepod fauna. Similarly, at the 

 beginning of December, north of South Georgia at Sts. 1054, 1056 and 1063 (Figs. 2«, 15), 

 the species was a somewhat less important constituent of the copepod plankton than in 

 this locality at the same time in the previous season. At Sts. 1054, 1056 and 1063 it formed 

 less than 50 per cent of the total copepod catch, but in 193 1-2 at St. 774 it formed more 

 than 75 per cent and at St. 775 about 50 per cent of the catch. It is not proposed here to 

 deal with the distribution of other species of Antarctic copepods, but it may perhaps be 

 mentioned that at Sts. 1019 and 1023, near the Falklands, the proportion of warm- water 

 species, characteristic of the sub-Antarctic zone, was higher than in the same region in 

 the preceding season (Sts. 750 and 751). Conversely, at St. 1029 (Fig. za) the proportion 

 of cold-water species, characteristic of the colder waters of the Weddell Sea, was higher 

 than at St. 788, which corresponds in position with St. 1029, in December 193 1. There 

 is a difference in date between Sts. 1019 and 750 of three weeks (9. xi. 32 and 30. xi. 31), 

 between Sts. 1023 and 751 of about a fortnight (16. xi. 32 and i. xii. 31) and between 

 Sts. 1029 and 788 of over a month (19. xi. 32 and 21. xii. 31). The higher proportion 

 of warm-water species between the Falklands and South Georgia in mid-November 1932 

 (Sts. 1019, 1023 and 1025), in contrast with that at the end of November and the be- 

 ginning of December 1 93 1 ( Sts. 750 and 751), may probably be correlated with the more 

 southerly position of the isotherms in 1932-3 than in 193 1-2 (Figs. 5, 6). As we have 

 already seen (p. 303) Deacon found the season 193 1-2 to have been colder and "later" 



