LIMITING FACTORS 189 



south-eastern Atlantic, may be cited. It is reasonably well established that Ceratium 

 fusus has its optimum, in the south, in the sub-Antarctic Zone, and if we examine the 

 extracts from Schimper's working notes (quoted by Karsten, 1905, pp. 34 et seq.) we 

 find a most significant distribution of this species. Proceeding south-westwards from 

 Cape Town, the ' Valdivia ' first encountered it in 41^° S at the surface. At the ' Valdivia ' 

 St. 120 in 42° 17' S it was present in addition in the deep haul (450 m.). A similar 

 vertical distribution was noted at the next station to the southward, but at St. 123 in 

 49° 07' S, 08° 40' E, i.e. well south of the Antarctic convergence in this longitude, the 

 species was present in the deep haul only. It will be realized that in the south-west 

 Atlantic, where Clowes' results indicate that the main component of this water is 

 easterly, this hypothesis would involve the replenishment of the plankton of the 

 Weddell Sea area, by the residue of the production in the north of the Bellingshausen 

 Sea. Thus there may be a continuous cycle of alternate vegetative and resting forms 

 travelling round the world on a course similar to that of a sailing ship running before 

 the westerlies and bringing the wind alternately on either quarter (supposing that the 

 direction of the wind remained constant). This would be in accordance with the fact 

 that the majority of Antarctic plankton forms appear to have an almost completely 

 circumpolar distribution, for it is in the relative proportions of the different forms, and 

 not in their presence or absence, that the populations of the various areas within the 

 Antarctic Zone chiefly difi'er. 



This hypothesis might explain many of the more marked fluctuations and anomalies 

 in the distribution of the plankton. 



Light is obviously a factor of the first importance in considering phytoplankton pro- 

 duction at any distance from the tropics. First Atkins (1928, p. 192) and latterly Marshall 

 and Orr (1930, pp. 870 et seq.) have maintained that it is the initial limiting factor, de- 

 termining the time of the start of the spring increase in north temperate waters. Gran 

 (1929 b, p. 50) held the view that off the Norwegian coast, the increase in inshore waters 

 began with the melting of the snows in spring, which washed down nutrient materials 

 from the land into the productive upper layers ; but the more recent work of Braarud 

 and Klem (1931, pp. 68 and 77) in the same area has shown that light is almost certainly 

 the initial limiting factor there also, and that the correlation between the main diatom 

 increase and the spring thaw is incidental, though it may lead to a smaller secondary 

 increase close inshore somewhat later {loc. cit., p. 69). The fact that the midsummer 

 cessation of phytoplankton production in temperate waters could be clearly demon- 

 strated to be due to the depletion of the nutrient salts appears to have led to an 

 over-estimation of their importance as a limiting factor at other times and in other 

 places. 



In the south it is certain that light is the limiting factor during winter, but this does 

 not account for the relatively late beginning of the spring increase off South Georgia. 

 November in the south is equivalent to May in the northern hemisphere and South 

 Georgia is situated in a latitude corresponding to that of the north of England. The 

 retardation must be due to a combination of adverse factors as yet imperfectly under- 



