PHYTOPLANKTON OF THE BELLINGSHAUSEN SEA 147 



antarctica. The absence of neritic forms here renders the conditions at St. 557 even more 

 difficult to understand. Again, at St. 602, a precisely similar though poorer phyto- 

 plankton was encountered even closer to the Biscoe Islands, the only obvious neritic 

 feature being the presence oi Biddiilphia polytnorpha in small numbers, and this despite 

 the presence of typical neritic associations both to the south (Sts. 598, 599) and east 

 (St. 601). 



Perhaps the richest haul of this survey in the Bellingshausen Sea proper was obtained 

 at St. 597, some distance to the north of Adelaide Island. The phytoplankton here was 

 of the Corethron — Rhizosolenia gracillhna type, and, as at St. 603, the proportion of 

 Corethron was high, much higher than at some of the stations worked later during the 

 previous year, when on some occasions Rhizosolenia gracillima was present in equal 

 numbers, or even in excess, in phytoplankton of this type. The next station dealt with 

 in Table XLIX furnishes the probable explanation. It will be seen that at St. 559 

 worked earlier in the year closer in to the land, a rather similar haul was taken, not so 

 rich in quantity, and with Chaetoceros neglectus and Frogilaria antarctica both present 

 in considerable numbers in addition to the two dominants. Now, in the eastern portion 

 of the Bellingshausen Sea, it seems that the first association to develop is one with 

 Corethron valdiviae and Chaetoceros neglectus dominant, succeeded at times (e.g. St. 602) 

 by almost pure Corethron. This may pass on to an association of the Corethron — 

 Rhizosoletiia gracillima type, which as we have already tried to show has been found most 

 abundantly late in the year in comparatively open water. It seems that the proportion 

 of Rhizosokfiia gracillima increases with the age of the surface water — up to a point. 

 Stations worked in Bransfield Strait always showed Corethron to be strongly dominant, 

 but others worked in the southern half of Drake Passage (Sts. 609, 610) showed the 

 two species in approximately equal numbers, as in the late hauls farther south in 1930. 



At St. 601 between the Biscoe Islands and the Fallieres coast, a neritic plankton 

 essentially similar to that found at the Palmer Archipelago was encountered. The more 

 numerous species were Corethron valdiviae, Chaetoceros neglectus, Thalassiosira ant- 

 arctica, Fragilaria antarctica, Nitzschia sp. (branching colonies), and Rhizosolenia 

 truncata, in that order of importance. Several of the rarer forms at this station were 

 never met with far from land, for example, Biddulphia polymorpha and Lycmophora sp. 



The very poor catch with Corethron valdiviae and Fragilaria antarctica dominant 

 obtained at St. 560 is possibly representative of conditions before the ice has dispersed 

 sufficiently for phytoplankton production to make much headway, as it was taken right 

 up against the heavy pack to the north and west of Adelaide Island on the outward 

 voyage. The last two stations analysed in Table XLIX were both made on the return 

 trip some three weeks later, close in to the land after most of the ice had cleared away. 

 At these stations, Sts. 598 and 599, a neritic association strongly resembling that found 

 at the Palmer Archipelago was observed, the dominant species being: Chaetoceros 

 tortissimus, Ch. neglectus, Corethron valdiviae, Fragilaria antarctica and Thalassiosira 

 antarctica. 



The analyses of the phytoplankton hauls from the stations worked between 66 and 



19-2 



