THE 'METEOR' RESULTS 271 



6-0° C. isotherm as the northern Hmit of his Antarctic zone, whereas we now know that 

 the highest surface temperatures reached by truly Antarctic surface waters (in the 

 hydrological sense) are of the order of 3-5° C. 



It is also important to remember that several of the small number of Antarctic 

 observations obtained by the Meteor were closer in to the land than the majority of our 

 own, and that the time was just after mid-season. This is just after the diatom maximum 

 in the northern part of the Antarctic zone, at a time when such dinoflagellates as are to 

 be found there will be at their maximum for the year. It may here be mentioned that 

 all available evidence goes to show that the dinoflagellates are essentially a warm-water 

 group of organisms, and that their maximum occurrence in higher latitudes, where the 

 seasonal changes in temperature are considerable, almost invariably coincides with the 

 period of maximum temperature for the region in question. 



Considering the Meteor results in respect of those groups for which our sampling 

 methods were known to be inadequate — ^the Coccosphaeriales and the small dino- 

 flagellates — we must now turn to the detailed figures published in Hentschel's earlier 

 work (1932, pp. 114-23). Taking only those stations which fall within the Antarctic 

 zone as defined in the light of more recent hydrological work, it becomes necessary to 

 omit five stations now considered as sub-Antarctic. From the remaining twenty-seven 

 observations at o or 50 m. Coccosphaeriales were recorded at nine only, five in the west 

 Antarctic and four in the ' east Antarctic ' regions. At only two of these stations, one 

 at South Georgia and one near the northern limits of the Antarctic zone in the open 

 South Atlantic, was the group of any real importance numerically. It is interesting to 

 note that the species Pontosphaera hiixleyi, long known to be the most important 

 member of the group in northern waters, was alone responsible for these figures. No 

 Coccosphaeriales were recorded at any of the more southerly stations in open water. 

 When the excessively small size of these organisms is taken into account, we may 

 therefore safely say that the Meteor results support our contention that our picture of 

 the main phytoplankton cycle in the Antarctic zone is unlikely to be affected by the 

 inadequacy of our methods for dealing with the members of this group. 



All writers on Antarctic phytoplankton have testified to the scarcity of Dinoflagellata 

 in those seas, but the Meteor was the first expedition to use methods capturing the 

 smallest ones in our area. Considered numerically therefore it is not surprising to find 

 the proportion of Dinoflagellata much higher than one was previously inclined to 

 suppose, particularly in view of the time of year at which the observations were obtained. 

 They averaged 15% of the total phytoplankton. Further examination of the Meteor 

 results reveals, however, that more than half (567%) of these were Gymnodinians 

 without chromatophores, and therefore presumably heterotrophic. Moreover, those 

 stations at which the numerical proportion of dinoflagellates to diatoms was high were 

 again very close in to the land. Another point to be borne in mind is that so far as is 

 known the division rate of dinoflagellates is considerably lower than that of diatoms. 

 We may say, therefore, that while minute dinoflagellates missed by our nets may be of 

 slight importance as producers during the post-maximal period for diatoms, it is 



