422 



DISCOVERY REPORTS 



demands explanation if the claim that it is essentially a sub-Antarctic species is to be 

 justified. The large proportion of empty frustules observed is in itself a strong indication 

 that round South Georgia the species does not find itself in a suitable environment, as 

 they are rarely seen where the species flourishes farther north. The suggestion put 

 forward in 1934 (p. 188) to account for the maintenance of the abundant flora of 

 the northern part of the Antarctic Zone still seems to furnish the only reasonable ex- 

 planation of its presence. I quoted the appearance of dead and dying individuals of 

 R. curvata and other sub-Antarctic species over the bank to the north-west of South 

 Georgia in support of the hypothesis that resting stages of truly Antarctic diatoms might 

 be transported southwards from the convergence region by the same agency, namely, 

 the warm deep water. It was thought that the appearance of the sub-Antarctic species 

 in the area indicated was due to subsurface transportation followed by vertical mixing 

 where the circulation was complicated by the bottom topography between South 

 Georgia and the Shag Rocks. After the examination of a much larger amount of 

 material, this explanation of the presence of R. curvata in the region in question still 

 seems to hold good, with the necessary modification that not all the individuals so trans- 

 ported are dead. Several of the South Georgia specimens appeared to be living, and 

 able to survive a short period at the height of summer. Most of them, however, looked 

 very unhealthy, and the absence of large individuals, such as would result from the 

 formation of auxospores, strongly suggests that the species is unable to persist in this 

 area for any length of time. 



Table I shows Hardy's records of R. curvata in South Georgian waters from material 

 collected during the summer of 1926-7. Although it is never a very abundant species, 

 even in its proper habitat, comparison with Table II clearly demonstrates its relative 

 scarcity in the South Georgian samples. 



Table I. Early observations round South Georgia 



These records may be taken as typical of the scanty occurrence of the species off 

 South Georgia, for none of the samples from that area examined subsequently contain 

 more than the maximum number, 1200 cells per net haul, recorded by Hardy from 

 St. 137. 



It is possible that the presence of this species at the ' Valdivia's ' most southerly 

 records was due at least in part to the same factor which is believed to account for its 

 presence round South Georgia. These records are from the South Atlantic north-east 



