RHIZOSOLENIA CURVATA 



439 



From this table it can at once be seen that the proportion of large individuals was 

 greatest in December-January and in May. In September also, although the large class 

 did not quite reach 50 per cent of the population, the increase in the proportion of large 

 individuals over the previous month was particularly noteworthy. This suggests that 

 auxospore formation immediately precedes or accompanies each increase in numbers of 

 the species, which, as has already been shown, occur in October, December-February, 

 and May. In February, when it seems certain that the greatest numerical abundance of 

 the species was reached, large individuals were not so common as in the preceding. If 

 the individual diameter classes for February are considered, it can be seen that the 

 largest was that composed of individuals of from 50 to 59/x diameter, just on the small 

 side of the mean, but that individuals of the smallest class were altogether absent during 

 that month. This strongly suggests that vegetative division was then at its maximum, 

 following maximal auxospore formation in January. The predominance of individuals 

 of the smallest classes in October and November, and in the winter months of June and 

 August, is well marked ; and it has been shown that at these times the species is either 

 decreasing or present in minimal numbers, but the result for the winter months were 

 derived perforce from data not strictly comparable with that given for the rest of the 

 year. 



The results of the attempted correlation between size and temperature are shown in 

 Table IX, based on figures collected in the manner described on p. 421. The proportion 

 of small individuals was highest at the lower temperatures, and again at the highest 

 temperatures of all. R. curvata was recorded from three stations with temperatures 

 above 8° C, but these did not furnish sufficient material to warrant the calculation of 

 percentages and are not considered in the table. It was noteworthy, however, that at this 

 extreme upper limit of the organism's temperature range, all the individuals measured 

 were below 50/j. in diameter. Large individuals predominate towards the upper limit of 

 the organism's temperature range but not near the extreme upper limit. This secondary 

 diminution in size at the highest temperatures that the organism can tolerate constitutes 

 the only point of difference from the size/temperature relation that one would expect by 

 analogy with Wimpenny's recent work (1936). 



Table IX. Size j temperature correlation in Rhizosolenia curvata 



Data not strictly comparable. 



4-2 



