THE LARVAL POPULATING OF THE EAST WIND-WEDDELL SURFACE STREAM 413 

 negative occurrences shown are based throughout on observations falling within the period November 

 to March, the northern spawning season, the number of such observations, for each successive 

 southern season in which, from 1926-7 onwards, both we and the Norwegian expeditions (p. 289) 

 visited the Antarctic, being shown per 30° of meridian by the coverage figures. The negative observa- 

 tions of the Mawson (B.A.N.Z.) expedition, which (p. 289) operated between Enderby Land and the 

 Ross Sea in the seasons 1929-30 and 1 930-1, although plotted in Fig. 145, were received too late to be 

 included in the coverage figures. They provide continuous cover, however, in the sectors 3o°-6o° and 

 60 "-90° E in the season 1929-30 and continuous cover between 60° and 180° E in the season 1930-1. 

 Thus, to take a good example, it will be seen that in 1937-8, at some time or another during the period 

 of northern spawning, we have observations covering, in all but one instance (between 0° and 30° E) 

 substantially, the entire circumpolar sea. 



It will be seen, then, that in all but two seasons (1927-8 and 1928-9) in which we worked these 

 southern waters from November to March, the principal concentrations of the surface larvae were 

 found massed in the western Weddell drift, the only substantial occurrences elsewhere being confined 

 to the coastal or near-coastal waters of the East Wind drift or, as for instance west of Graham Land, 

 to water evidently of East Wind origin. There can be little doubt, then, that the summer outburst in 

 Weddell West is an annual event as no doubt also are the coastal risings that take place in the East 

 Wind drift. Our failure in 1927-8 and again in 1928-9 to strike any major risings in the western 

 Weddell drift is almost certainly due to inadequate sampling. In both seasons the main east-flowing 

 part of Weddell West where they are so abundant was not examined, the observations being confined 

 to the South Georgia and Bransfield Strait whaling grounds where the summer larvae have repeatedly 

 been shown to be scarce or absent. 



It can be shown too (Fig. 146), from a similar marshalling of the data, that, following the summer 

 outburst in Weddell West, the April to November massing of the older larvae throughout the 

 Weddell stream to its farthermost limits, but no farther, is also an annual event. There must 

 too it seems be annual, if sporadic, intrusion of larvae from the East Wind into the West Wind 

 zone. At any rate on each of the four annual visits we paid to the Kerguelen area larvae were en- 

 countered, on three of them, as Fig. 146 shows, in some substantial measure of abundance. The 

 annual position in the East Wind tongues, however, is rather difficult to judge, our crossings of 

 these circumscribed regions of larval occurrence being infrequent and our coverage for the most 

 part indifferent. 



LOCAL DISTRIBUTION. THE SOUTH GEORGIA WHALING GROUNDS 



On our circumpolar charts the distribution of the krill in such a circumscribed and closely sampled 

 region as this obviously cannot be shown in detail and accordingly this section is presented partly to 

 fill this want and partly to show how the larvae in the Weddell stream get carried towards this island 

 in the surface drift and how both they and the older stages to which they give rise subsequently 

 become distributed in its surrounding waters. I do not, however, propose to deal with this matter 

 at length since the South Georgia charts are largely self-explanatory and a detailed discussion of them 

 would involve the repetition of much of what has already been said in the long section presenting the 

 distribution on its circumpolar scale. 



The influence of the Weddell current in so far as it aflFects this locality, following Deacon (1937, 

 p. 19, Fig. 4), may be described in broad outline as follows: 



Under the influence of the Scotia Ridge part of the main east-flowing Weddell stream is diverted 

 north-westwards towards South Georgia and as it approaches the land divides into two tongues 



