BIOGEOGRAPHICAL REGIONS AND AREAS 281 



Kerguelen Island and over the ridge connecting it with Heard Island may be expected 

 to resemble those observed in the South Georgia area on a smaller scale, but we have 

 no observations there. 



The subdivisions described are shown in Fig. 2, and may be tabulated as follows : 



Main Regions (oceanic) 



The Northern Region: between the Antarctic convergence and a line 330 miles south of it, all 

 round the world, excepting the special areas between 30 and iio°W, and between 150° W and 

 170° E. 



The Intermediate Region : between the southern limit of the above and the Antarctic circle all the 

 way round the world with the exception of the same complicated areas. 



The Southern Region: all seas south of the Antarctic circle, excluding immediate coastal areas. 



Special areas 



The South Georgia area: between 52 and 55° S; 33 and 41° W. Neritic influence very strong. 



The Scotia Sea: between the Antarctic convergence and 62° S : 30 and 70° W, excluding the South 

 Georgia area. Neritic influence considerable but less marked. 



Other Special areas: where our observations are too few for detailed consideration, namely: 

 (i) The eastern south Pacific between the Antarctic convergence and the Antarctic circle: 70- 

 110° W. This is essentially oceanic and is best known. (2) The area north of the Ross Sea between 

 the Antarctic convergence and the Antarctic circle: 150° W-170'' E, oceanic. (3) Central Weddell 

 Sea between the southern limits of the Scotia Sea and the Antarctic circle, oceanic. (4) Bransfield 

 Strait and coastal waters of the Palmer Archipelago, neritic. (5) Other essentially neritic areas, e.g. 

 coastal waters of the Balleney Islands, which could be ranged according to latitude if necessary. 



It will be seen that the main idea of this scheme of subdivision is essentially similar 

 to that which I had already suggested to Clowes (1938, p. 8), but with three times as 

 much data it has been possible to improve the original zonation. The definition of the 

 southern region (or zone) in terms of distance from the ice-edge has been abandoned 

 for the arbitrary one, placing its northern limit at the Antarctic circle. This is an im- 

 provement in one way because of the difficulty of establishing an ' average summer 

 position' of the ice-edge in the less known sectors, but it is certainly true that the 

 actual extent of the pack-ice is a most important environmental factor in this region. 

 It has also been possible to define the special areas whose existence had indeed been 

 recognized though it was not possible at that time to express that recognition in con- 

 crete terms. In all other respects it will be seen that the scheme remains essentially 

 the same as that which Clowes found helpful in considering the distribution of phos- 

 phate and silicate in the water. This in itself provides evidence that it has real signifi- 

 cance despite the unavoidably arbitrary nature of some of the boundaries. 



ECOLOGICAL GROUPING OF THE IMPORTANT 

 PHYTOPLANKTON SPECIES 



In considering the phytoplankton population in such a vast region as the Antarctic 

 zone, it is obviously desirable to adopt some scheme of ecological characterization of the 

 important species. By such means only can the bulk of observational data be clarified 



