ANTARCTIC SURFACE WATER 47 



SEASONAL VARIATION OF SILICATE IN THE SURFACE LAYER 

 AROUND SOUTH GEORGIA 



At present there are insufficient data for a complete examination of the seasonal 



variation of silicate in the surface layer at South Georgia. There are observations in 



December 1932 (twelve stations), February 1933 (four stations), November 1933 



(twelve stations), August 1934 (twelve stations) and February 1935 (twelve stations), 



but no data from the critical months of October, January, April and May. The importance 



of securing data from these months is that the steepness of the fall of silicate content 



in early spring is hidden by the absence of October values ; the probable position of the 



[l?] 1934 



O = 0-100 METRE5 AVERAGE 

 ® = 5URFACE 



O [4-] 1933 

 ® [4] 1933 



SEPT OCT NOV DEC JAN FEB MAR 



Fig. 11. Possible curves for the seasonal variation in silicate content of the surface and of the average of the 



0-100 m. layer at South Georgia. 



minimal value close to 1 January also cannot be shown, and any evidence for an autumnal 

 secondary outburst of phytoplankton is missing. The available figures have, however, 

 been graphed and are shown in Fig. 1 1 as possible curves for the average contents of 

 the surface and 0-100 m. layer between late August and mid-February. The figures in 

 parenthesis opposite the points on the curves refer to the number of observations available. 

 As may be seen, the average value of the 0-100 m. layer falls slowly from a value of 

 about 1850 mg. at the end of August until about 1 November, when the gradient becomes 

 much steeper and the silicate content falls to a probable minimal value of less than about 

 400 mg. near 1 January. Regeneration must be fairly rapid because in late February 

 1933 the 0-100 m. content was 550 mg., and in early February 1935 as much as 950 mg. 



