332 DISCOVERY REPORTS 



parison the period of the main increase has been taken as from the date of the first 

 clear increase over the minimal winter values to the first pronounced descent towards 

 the post-maximal decrease. These dates naturally differ in the several areas, and 

 the figures have, therefore, been tabulated below in addition to the diagrammatic 



representation. 



Table 12 



It will be seen that in the three oceanic areas the values are much the same and 

 slightly higher than in the English Channel. In the Southern Region, where the period 

 of the main increase is much shorter (and where there is no secondary autumnal increase), 

 the total production will, of course, be much smaller. Off South Georgia and in the 

 Scotia Sea the much higher values correspond with the relative degree of neritic 

 influence in these two areas. 



The seasonal cycles described clearly support the views already put forward (Hart, 

 1934, p. 193) that the physical factors of the environment play the most important part 

 in determining the course of phytoplankton production within the Antarctic zone. Most 

 important are : light, the degree of stability of the surface layers, and the interrelated 

 effects of pack-ice. These are certainly prime causes in the determination of the time 

 of the onset of the main increase, and the extent and duration of the autumnal secondary 

 increase in the more northerly parts of the Antarctic zone. However, they do not by 

 themselves explain the post-maximal summer decrease in the more northerly Antarctic 

 surface waters, or the vastly greater production in neritic as compared with oceanic 

 areas. Since physical factors alone do not sufficiently account for these features, their 

 probable explanation must be sought among chemical and biological factors. 



From earlier work we know that while decrease in phosphate content of the surface 

 waters may augment the post-maximal decrease in phytoplankton, it is extremely 

 unlikely that shortage of this nutrient salt is ever sufficient to account by itself for that 

 decrease (Hart, 1934, p. 184; Clowes, 1938, p. 112). So far as the scantier evidence 

 goes, the same may be said of nitrate. The reduction of silicate, on the other hand, is a 

 very probable cause of the post-maximal decrease in the almost purely diatomaceous 

 phytoplankton with which we are concerned, as has already been suggested hypothe- 



