ANTARCTIC SURFACE WATER 39 



the end of December. In this connexion it is remarkable that a rise takes place 

 in the southern curve during November, December, January, and Februarv during 

 which time the southern region phytoplankton is slowly and steadily coming to a 

 maximum in the third week of February. If the minimum of i November in the curve 

 of the southern region is a reflection of the minimum in the northern curve at the 

 end of the previous December, then the rise in the silicate content in the southern 

 region for the months of November, December, January and February should be related 

 to the rise in the northern curve for January, February, March and April. If the two 

 minima are thus correlated the conditions in the southern region depend on those 

 occurring 10 months or possibly i year and 10 months before in the northern region. 

 It is possible that this will give some basis for a speculation as to how long silicate takes 

 to be returned to the Antarctic. 



As a summary of our present knowledge of the silicate content of the surface layer 

 in the Antarctic zone the following remarks may be of service. The silicate content of 

 the o-ioo m. layer in two arbitrarily defined regions of the Antarctic zone when 

 plotted as 2-monthly averages give interesting curves. The rises and falls of these two 

 curves are partly at variance with known dates and effects of phytoplankton growth 

 and decline in situ. It is suggested that because the amount of silicate which leaves 

 the Antarctic zone exhibits a seasonal change, the return cycle of silicate will cause 

 this seasonal change to be mirrored in the silicate content of the surface layer in the 

 southern region. Also, the northern region averages reflect the past history of phyto- 

 plankton activity in the southern and intermediate regions. 



SEASONAL VARIATION OF PHOSPHATE AT THE SURFACE 



The averages of phosphate and silicate contents of the o-ioo m. layer in two arbitrarily 

 chosen regions of the Antarctic zone have been presented in the form of curves of 

 2-monthly means which gave an insight into the seasonal variation of these nutrient 

 salts in the surface layer at the extreme south and north of the zone. The following 

 remarks are a summarized account of the surface values in different months of the year 

 as far as our observations allow. It is difficult to make a concise summary because the 

 main phytoplankton outburst varies in date and extent. There is also some evidence 

 of a small autumnal increase of phytoplankton, even more unevenly distributed. The 

 surface and o-ioo m. layer observations for special areas, where data are more 

 numerous, will be given later. 



The average surface phosphate content at the ice-edge in September is of the order of 

 125 mg., except north-east of the Ross Sea where a higher value of over 140 mg. was 

 recorded in 1932. The northern limit of the Antarctic zone has an average of about 

 no mg., again except north-east of the Ross Sea. 



In October surface values at the ice-edge of about 110-115 mg. have been found, 

 whilst northern boundary values of about 105 mg. are recorded. 



In November ice-edge surface values are about 105- no mg. whilst northern boundary 



