426 



DISCOVERY REPORTS 



abundance in late summer or autumn, after the main diatom flowering has passed its 

 peak. Examples are R. styliformis in the northern hemisphere, and R. polydactyla, 

 R. chunnii and R. antarctica (somewhat farther south) in the southern. This may be in 

 some way connected with the variation in intensity and duration of light, for the other 

 environmental factors are vastly different in corresponding latitudes in the two hemi- 

 spheres. 



OBSERVATIONS IN THE SOUTH ATLANTIC OCEAN 



Our findings of R. curvata in the South Atlantic Ocean east of South Georgia are 

 shown in Fig. 2, while particulars from each station will be found in Table III below. 



Table III. Observations in the South Atlantic Ocean 



Unfortunately our data from this area are more scanty than from any other, apart 

 from the middle of the Southern Indian and Pacific Oceans. As will be seen later, the 

 observations are well distributed over the greater part of those sectors, but they are not 

 so here. The outstanding feature of these records is the relatively large numbers of the 

 species obtained at some stations south of the Antarctic convergence, while relatively 

 poor catches were obtained to the north. This is at variance with the claim that the 

 species reaches its optimum in the sub-Antarctic Zone. It will be readily seen from 

 Table III, however, that not only were all the numbers recorded below the average, 

 but that the sub-Antarctic records from this area were, with one exception, obtained 

 during the months when the species has never been found abundantly. These facts go 

 far towards explaining the apparent anomaly, for the relatively abundant southerly 

 records were obtained at different times of year, when the data from the sub-Antarctic 

 Zone in this sector are inadequate. 



The unusual southward extension of the species at Sts. 1052-1054 was almost cer- 

 tainly due in part to mixing from across the convergence, which is known to be par- 



