RHIZOSOLENIA CURVATA 



437 



evidence as to the meaning of its observed distribution. Wimpenny's work on R. 

 styliformis in the North Sea suggests that auxospores can only be formed near the upper 

 limits of the organism's temperature range. It was for this reason that the measurement 

 investigations shortly to be described were undertaken. With the species limited to these 

 two methods of reproduction it is obvious that it cannot persist in an environment in 

 which auxospore formation is impossible, though it may survive for a time by continued 

 binary fission, with the cells getting ever smaller. Therefore if it could be shown that 

 smaller individuals were present in marked excess at the cold stations south of the con- 

 vergence, while larger individuals that could only have been the result of recent auxo- 

 spore formation were relatively abundant at the warmer stations to the north, Wim- 

 penny's generalization could be extended to this species, and lend powerful support to 

 the view that the sub-Antarctic Zone is its proper habitat. 



SEASONAL VARIATION IN ABUNDANCE 



Owing to the limitations of the data available, it is not possible to establish the 

 seasonal variation in abundance of R. curvata with the certainty possible when dealing 

 with common species from intensively studied areas. It was essential to endeavour to 

 obtain some idea of the probable seasonal range in numbers, in order to be able to assess 

 the value of the distributional variations observed. The available records have been 

 grouped into month classes, and arithmetical means for each month calculated as 

 follows : 



Table VII. Seasonal variation in abundance of Rhizosolenia curvata 



In this table the months have been arranged so that July, corresponding seasonally to 

 January in the northern hemisphere, comes first. The more obv< us features of the 

 seasonal variation indicated by these means agree so well with our general knowledge 

 of plankton periodicity in the south, and the genus Rhizosolenia more particularly, that 

 it seems certain that a useful approximation to the true order of frequency has been 

 arrived at. Thus the low values recorded in the winter months June and August, when 

 conditions for diatom growth are known to be very unfavourable, are in accordance with 

 expectation. The slight increase in October, corresponding to April in the northern 



