lo DISCOVERY REPORTS 



south of Elephant Island, but north of the South Orkneys. In the season 1930-1 it 

 continued as far as the south-western coasts of South Georgia, but in normal seasons, 

 except perhaps at the equinoxes, when the flow is strongest, it probably loses its identity 

 some distance to the south-west of the island, becoming merged with old water from 

 the Bellingshausen Sea. This type of Weddell Sea surface water is here provisionally 

 termed the '"ivestern" Weddell Sea water, in order to distinguish it from the "eastern" 

 Weddell Sea water next to be considered. The two types can be readily distinguished 

 by the phytoplankton which they contain. 



The eastern Weddell Sea water enters the sea in comparatively low latitudes. It 

 does not penetrate so far into the Weddell Sea proper, nor remain there so long, as 

 water of the western Weddell Sea type. On its north-easterly course when flowing 

 out of the sea, it remains to the south of the western type water, and continues clearly 

 recognizable by the character of its phytoplankton as far as the South Sandwich Islands. 

 Here the arcuate ridge upon which these islands are situated apparently causes a north- 

 westerly deflection of this water, which has been detected on several occasions in the 

 form of a cold tongue projecting some distance along the north-east coast of South 

 Georgia, at a distance of some 50 miles out from the island. 



The individuality of the two types of surface water in the Weddell Sea was first 

 clearly recognized by Mr D. Dilwyn John and his colleagues in the R.R.S. ' Discovery II ' 

 in 1932. It has been found to afford a valid explanation of the observed diff'erences in 

 the phytoplankton material collected in previous seasons. Mr John was able to divide 

 the surface water into four types, the very cold water far to the south, water containing 

 mehing pack-ice, that in which pack-ice had recently melted, and "older" water. These 

 types were found to be characterized by a certain definite range of surface temperature, 

 and by the nature and bulk of the phytoplankton they contained. These limits probably 

 apply only to the time of observation, mid-season 193 1-2, as the range of seasonal 

 variation is considerable. Nevertheless, this conception of the "age" of the surface 

 water is of the first importance, and, generally speaking, Mr John's preliminary results 

 agree very closely with those obtained at mid-season in previous years. One important 

 difference between surface waters of the Bellingshausen and Weddell Seas was noted, 

 that in the Bellingshausen water the heaviest phytoplankton hauls were obtained in 

 water containing melting pack-ice, while in the Weddell Sea water the richest hauls 

 were obtained in the old water which had travelled farthest from the pack. This is in 

 good agreement with our mid-season results, but apparently does not hold earlier in 

 the year, when very heavy catches have been obtained in Weddell Sea water, south of 

 South Georgia and close up to the edge of the pack. 



The ice conditions in the seas with which we have to deal are notoriously variable, and 

 little is definitely known concerning them even by the seamen who use these waters 

 most. Work in this direction is in progress, and when published should materially assist 

 in explaining the distribution of the plankton. It is evident that some definite relation 

 between the phytoplankton production and the position of the ice-edge exists, and the 

 possible nature of this is discussed later. 



