362 DISCOVERY REPORTS 



(3) The late Furcilias, Furcilias 4-6, become dominant in the latter half of March, those thus 

 encountered then evidently stemming from the earliest January risings. 



(4) As expressed by the stage frequency, the modal values of the summer swarms exhibit a 

 pattern that becomes increasingly heterogeneous as the season advances. 



{b) Southern or East Wind zone 



In these high latitudes a vastly different summer condition is disclosed, the First Calyptopis 

 it would appear persisting as the single dominant surface stage right through to the end of 

 summer. 



Autumn. The distribution and relative abundance of the massed surface larvae in autumn (April- 

 June) is illustrated in Fig. 112, which shows how the dense summer swarms, originally concentrated 

 principally in Weddell West (Fig. no), have spread eastwards and, in equal density, are now 

 scattered along the surface stream from end to end. Some of them, too, have reached the South 

 Georgia whaling grounds (in May) and some (in April) the Bransfield Strait. In the latter locality, 

 however, it is likely that the April swarm recorded (p. 364, Fig. 113, Station 198), in view of 

 its position at the western end of the channel, drifted in from the Bellingshausen Sea and, like the 

 April swarm of similar age encountered in the Drake Passage (p. 315, Fig. 78, Station 383), 

 might, it seems, be traced back to the northward flow of East Wind water that occurs near Peter I 

 Island. 



The observations in the West Wind drift, particularly those between 30° and 75° E in April and 

 May, suggest again (p. 320) that the surface larvae, already spread densely over the whole of the 

 Weddell drift by April, do not in fact get carried far beyond 30° E except as occasional stragglers. 

 Elsewhere, the West Wind zone is barren except for three isolated and widely separated instances 

 where deflection of cold water from the East Wind drift has evidently carried with it moderate to 

 substantial larval pockets into the warmer east-going drift in the north. These intrusions from the 

 south, ^ in their isolation, appear conspicuously in Fig. 112, namely, south-east of Kerguelen, an April 

 record, to the north of the Balleny Islands, a June record, and north-east of Peter I Island, also a June 

 record. All (p. 360) probably first made their appearance in the West Wind drift some time earlier or 

 later in the summer, somewhere to the south, or more likely perhaps to the south-east, of where they 

 were severally recorded. Since they would then, however, have been much younger, almost certainly 

 in fact in the First Calyptopis stage, they would not have been detected by the stramin nets. The 

 advanced state of their development relative to that of their contemporaries in higher latitudes (Fig. 87) 

 is discussed on p. 327. 



In the Pacific sector our autumn observations are rather few and scattered. Such as we have, how- 

 ever, are negative and having regard to the position there in summer, and later, when we again have 

 substantial observations (p. 369, Fig. 117), in spring, there can be little doubt that the entire 

 region can rarely if ever be populated by the larval forms. 



By April, as Fig. 112 shows, the East Wind drift becomes largely inaccessible owing to the 

 northerly encroachment of the winter pack. In the only part of it in which we have observations, 

 however, namely, between 0° and 60° E, a moderate to substantial surface population is revealed, 

 the richest near the continental land. 



The autumnal developmental condition of the principal concentrations of the massed surface 

 swarms is shown in Figs, in and 113, the former covering such small part of the East Wind zone 

 as we were able (Fig. 112) to examine, the latter the whole of the now richly populated Weddell 

 stream from west to east and including seven stations located in the northerly extremities of the 

 several East Wind outstreamers referred to above, and one (Station 198) from Bransfield Strait. 

 * See also the earlier charts of the larval distribution for March and April. 



