48 DISCOVERY REPORTS 



was present. If the proposed values for the maximal and minimal contents around 

 South Georgia are correct then the net withdrawal of silicate in the o-ioo m. layer is 

 about 81 per cent of the available silicate (see, however, footnote on p. 112). The surface 

 withdrawal is most certainly of the same high order. 



SEASONAL VARIATION OF PHOSPHATE IN THE SURFACE LAYER 



IN THE SCOTIA SEA 



The Scotia Sea is bounded by the rise of the sea bottom in the arc extending from 

 the Burdwood Bank towards the Shag Rocks and South Georgia and continued through 

 the South Sandwich Islands, bending back to the South Orkney and Shetland Islands. 

 We have seasonal data from two main lines of stations running northwards across the 

 sea, one from Elephant Island to the Falklands and the other northwards from the 

 South Orkney Islands across the Antarctic convergence. In addition we have data from 

 a line of stations (Sts. 633-638) midway between the other two lines. It was considered 

 that the few stations on these lines which lie north of the Scotia arc should also be 

 included in this discussion owing to the common origin of the water at these stations 

 and those in the Scotia Sea proper. 



The surface and 0-100 m. layer averages have been plotted for these lines of 

 stations, and are shown in Figs. 12, 13. For the stations northwards from the 

 South Orkney Islands Fig. 13 shows a very uniform phosphate content from south to 

 north in the months of October and January at the surface and in the 0-100 m. 

 layer. In January the Antarctic surface water is universally low in phosphate content 

 and the curves show the effect of the main phytoplankton outburst. The surface 

 value is as low as 71 mg. and varies between 71 and 79 mg. in the Antarctic zone. 

 In October the phosphate content is again uniform across the sea and represents the 

 conditions before the main phytoplankton outburst which occurs a short time later. 

 The phosphate content in October is of the order of no mg. at the surface and is 

 not very much greater in the 0-100 m. layer, a fact which shows the effect of winter 

 mixing. The October values are probably not maximal, and in the absence of July or 

 August figures we can only say that the net amount of phosphate which is withdrawn 

 from the surface by the phytoplankton during the spring and summer season is at 

 least 36 mg. 



The April curves for the stations north of the South Orkney Islands and the March 

 curves for the line a few degrees to the west are interesting in their lack of uniformity 

 when compared with those of January and October. The phosphate content at any 

 position is merely the reflection of some past concentration of phytoplankton and of 

 regeneration or return of this salt. In April the curves for both the surface and the 

 0-100 m. layer show wide variation, the surface value varying between the low 

 values found in January and the high contents of October. Regeneration must be 

 proceeding in April, but the effect of regeneration may be masked by changes in 

 movements of the surface water, by localized autumnal secondary outbursts and by the 

 arrival of water which has experienced peak conditions farther south and is now low 



