112 



DISCOVERY REPORTS 



corner. The fine bathymetric chart based on some thousands of echo-soundings (Herd- 

 man, 1932, p. 231, pi. xlvii), favours this view as it shows clearly that the deeper inlets 

 are all in the direction one would expect from the plankton hauls. 



The obvious features of this survey were the very pronounced dominance of C. 

 valdiviae to the exclusion of almost all other phytoplankton forms, the small number of 

 species recorded, and the absence of any marked effect from the eddy of Weddell Sea 

 water round Joinville Island on the phytoplankton. The hydrological results, however, 

 showed that this eddy was still present, if not so clearly distinguishable as on subsequent 

 surveys worked earlier in the season. The first two of these three main features are well 

 shown by the following list of all the species recorded during the survey, arranged in 

 order of their estimated total abundance, and showing the number of stations at which 

 each occurred : 



Corethron valdiviae (18) 

 Thalassiosira antarctica (10) 

 Chaetoceros criophilum (11) 

 Rhisosolenia alata f. gracillima 

 Nitzschia seriata (8) 

 Fragilaria antarctica (4) 

 Chaetoceros atlanticus (2) 

 CIt. neglect us (i) 

 Ch. schimperianus (i) 

 Ch. dichaeta (i) 

 LeptocyUiidrus sp. (i) 

 Synedra spathulata (4) 



83,493,600 Rhizosolenia styliformis (4) 



387,450 Silicoflagellata (i) 



365,000 Dactyliosolen laevis (i) 



(11) 202,800 Synedra sp. {1) 



183,000 Thalassiothrix antarctica (2) 



141,400 Biddulphia striata (2) 



71,700 Eiicampia antarctica (2) 



58,800 CoscinodisciiS spp. (i) 



29,400 Periditiium antarcticum (i) 



25,200 Dactyliosolen antarcticus (i) 



16,800 Corethron valdiviae (spineless chains) 



16,500 Lycmophora sp. (i) 



(0 



16,350 



12,600 

 8,400 

 8,400 



7,800 



6,000 



4,200 



3,000 



3,000 



600 



300 



300 



Estimated total phytoplankton 85,081,600. 



From this it will be seen that Corethron valdiviae dominated over all the other species 

 to such an extent that it formed over 98 per cent of the phytoplankton. 



The distribution of Corethron valdiviae itself is indicated in Fig. 51. By comparison 

 with Fig. 50, which gives the phytoplankton totals, it will be seen that it was completely 

 dominant at all stations except the three in the extreme north-west. The richest hauls 

 of this survey, in the middle of the strait and at St. WS 383 to the north-east, consisted 

 almost entirely of this species. On subsequent surveys the richest hauls were obtained 

 farther to the east, so that it was fairly evident that on this occasion the phytoplankton 

 was being renewed by the influx of Bellingshausen Sea water into the western end of the 

 strait. The work carried out in the north-east of the Bellingshausen Sea itself during 

 subsequent seasons showed that a considerable increase of Corethron took place fairly 

 close in to Adelaide Island in early autumn, while farther to the north-east the remnant 

 of the more varied phytoplankton of the earlier months no doubt persisted. This may 

 furnish an explanation of the conditions found in Bransfield Strait on this occasion, as 

 it is clear from the distribution charts that little but Corethron was entering through the 

 broad, comparatively shallow gap between Low and Hoseason Islands, while a more 

 varied phytoplankton was encountered in the narrower, deeper channels to the north- 

 west. It would seem reasonable to regard this as the resultant of the north-easterly drift 

 from what might be termed the extreme north of the Bellingshausen Sea in January, 



