PHYTOPLANKTON OF BRANSFIELD STRAIT 135 



continued importance of the smaller species mentioned above, and by the much larger 

 variety of species than are present in the old Bellingshausen Sea water. Another notable 

 feature is the much smaller proportion of Corethron valdiviae than elsewhere in the strait 

 at this season, and the absence of auxospore formation of this species, whereas in the 

 rich Corethron plankton drifting to the north-east with the old Bellingshausen Sea 

 water, auxospore formation is proceeding apace at this time. But little phytoplankton 

 is present in the water entering the western part of the strait at mid-season, probably 

 owing to the lateness of the spring increase in the ice-bound Bellingshausen Sea whence 

 this surface water originates. 



Still later in the year a moderately rich phytoplankton, with Corethron very strongly 

 dominant, and Rhizosolejiia gracillima the only other form of constant occurrence, de- 

 velops in the middle of the strait as a result of the invasion of Bellingshausen Sea water 

 from the westward. Up to the time when it enters the labyrinth of islands, and com- 

 paratively shoal water which guards the western entrance, this Bellingshausen Sea water 

 possesses a much more varied flora, which leads one to suppose either that these other 

 species fail to survive the violent vertical mixing consequent upon the passage of the 

 current through the various obstructions, or that conditions within the Bransfield Strait 

 itself at this time may be inimical to their extensive development. So far as they have 

 been studied, the chemical and physical properties of the surface water within the strait 

 do not, however, present any features sufficiently striking to account for such a big 

 diflference. The first of the two factors mentioned is thus more likely to be responsible 

 for the peculiar nature of the phytoplankton in Bransfield Strait. While other species, 

 such as Thalassiosira antorctica, penetrate in small numbers to the extreme north-west 

 (where the deepest opening, between Smith and Snow Islands, lies), nothing corre- 

 sponding to the pronounced dominance of Corethron valdiviae over the greater part of 

 Bransfield Strait throughout the year has been observed in any other area. In the present 

 state of our knowledge the most probable reason for this phenomenon is that a large 

 number of the other species fail to survive the violent vertical mixing consequent 

 upon the passage of the Bellingshausen Sea drift through the chain of shoals and islands 

 guarding the western entrance. Direct evidence of this mixing is furnished by the high 

 surface salinities at some of the stations in the gaps (e.g. at Sts. 607 and 608 already 

 quoted). The surface water to the south-west coming up from the Bellingshausen Sea 

 rarely exceeds 33-90, whereas at these two stations the salinities were 34-17 and 34-24 °l^^ 

 respectively, at the surface. As a definite set from the westward is a well-established fact 

 (vide Antarctic Pilot, 1930, p. 72) it follows that considerable mixing with the more 

 saline, deeper-seated water layers must have taken place. 



By early autumn the strongly defined character of the phytoplankton within the area 

 affected by the eddy of Weddell Sea water has entirely disappeared, though traces of 

 Weddell Sea inffuence are still clearly apparent in the hydrological features of the surface 

 water. The reason for this lies probably in the rapid waning and change in character of 

 the phytoplankton after mid-season, a feature that has also been noted at certain stations 

 between Clarence Island and the South Orkney Islands. 



