DISTRIBUTION OF PHYTOPLANKTON 



and January, and giving place to other species of Coscinodiscus and to Corethron val- 

 diviae. It was all but absent in the five Sts. WS i io-i 14 taken in autumn, late May, on 

 the C line ; 400 (4) only being taken at St. WS no. Our survey of December and January 

 appears to show Coscinodiscus bouvet in the act of declining and sinking. The majority 

 of them appear to die. It is interesting to speculate whether some of them remain 

 alive, perhaps forming resting spores, and sink from the surface layers of water, which 

 are moving northward, into the lower masses of water moving back towards the pole. 

 This might explain how species can maintain their population in the south when 



I 



o< 10.000 



□ 10,000-30,000 



rR 



: T%^f-^- 





B 

 3 



E 



B 



" 



3 



3 



U 



JO' *0° *y 39' JO' 38° JO' 37° JO' 36" JO' 



Fig. 16. Distribution of Coscinodiscus bouvet round South Georgia in the December-January 1926-7 survey. 

 Scale of squares represents numbers per haul with N 50 V net from 100 m. to the surface. (See section on 

 significance of plankton numbers, p. 40.) Negative observations shown as dots. The dotted lines BB and 

 WW mark the probable boundaries of the surface waters from the Bellingshausen and Weddell Seas re- 

 spectively, the zone in between being an area of mixing (cf. Fig. 6). 



they are continually being carried away from the pole in the surface layers. No such 

 spores were seen, but a few cells were found to be living even below 1000 m. All those 

 that were found in the top 100 m. appeared to possess chromatophores ; but below 

 this level the percentage of those containing chromatophores decreased as the depth 



