52 



DISCOVERY REPORTS 



the surface and the o-ioo m. layer shows very little variation from south to north across 

 the Scotia Sea in the months of January and October. The April observations fall mid- 

 way between those of January and October, showing that regeneration is having an 

 effect in the increased value of the April figures over those of January. The April values 

 for silicate content are more uniform than those for phosphate, and, except close to the South 

 Orkney Islands, the silicate content falls gently from south to north. The reason for this 

 may lie in the quicker regeneration in situ of silicate over phosphate, which latter is 

 partly held by the zooplankton, whereas a large proportion of the silica skeleton of the 

 diatoms is excreted when the phytoplankton is consumed by the zooplankton. The high 



AC 



MAR [{ SEPT 



LATITUDE 

 STATION 



60" S 



5B" 





m 



(VI 



1/1 in 



ttj OJ 



_l i_ 



54° 



52°S 



OJ 



aooo- 



q 

 in 



& 1 000 



O MARCH 1934- 



X SEPTEMBER 1934 



0-IOOm 



SURFACE 



&! c ? su RFACe 



x- - ~_~_-JT<X 



SEPT SURFACE*" 



SEPTEMBER 

 SURFACE and 0-100 



MARCH_0^120_-G 

 MARCH SUR? ACE ' 



Fig. 15. Surface silicate content and the average silicate content of the o-ioom. layer between the Falkland 



Islands and Elephant Island. 



value at the most southerly station in April is due to the presence of Weddell Sea water 

 close to the South Orkney Islands. The January values probably approach very closely 

 to the minimum concentration of silicate in the surface layer of the Scotia Sea, and, 

 in the absence of observations earlier than October, we can say that a minimum of 

 about 50 per cent of the available silicate is withdrawn during the phytoplankton season 

 at the south of the Scotia Sea, and something approaching 80 per cent in the north. 



Between the Falkland Islands and Elephant Island there are only two sets of data, 

 one obtained in March 1934 and the other in September 1934, each set of observations 

 having been made immediately previous to those in April and October north of the 

 South Orkneys. They are shown in graphical form in Fig. 15. Considerably less silicate 

 exists in the west of the Scotia Sea than north of the South Orkneys. In general the 

 September values are greater than the March ones and must approach the maximum 

 values in this position during the year. Considerably more data are required on this line 

 of stations before seasonal variation in the western part of the Scotia Sea can be discussed. 



