3o8 DISCOVERY REPORTS 



levels, not necessarily however (p. 304) directly below, of hatching eggs, then perhaps the overriding 

 inference to be drawn from Fig. 72 is that the major hatchings from which the krill population springs 

 take place (i) in Weddell West, especially between December and March, and (2) in the coastal or 

 near-coastal waters of the East Wind zone from the Ross Sea westwards to the Princess Martha Coast 

 between January and April. 



The shallow living First, Second and Third Calyptopes 



November-December. The larvae of this group have now completed the developmental ascent, and 

 as Fig. 73 shows in November-December some have already reached the surface, although so far 

 only in small numbers. The occurrences, such as they are, are mainly December ones, there being 

 only two in November, one of a Second Calyptopis on the South Georgia whaling grounds and the 

 other of a First Calyptopis (a Norwegian record) in Weddell Middle. Both are probably highly 

 exceptional. It will be seen that the two December occurrences of any note are located in the oceanic 

 water of the western Weddell drift^ more or less directly over the region where in the same month the 

 first deep stragglers from the rising Naupliar and Metanaupliar population were recorded. Elsewhere, 

 apart from a few widely scattered Calyptopis stragglers, there is no indication yet, that is, in December, 

 of any large-scale accomplishment of the developmental ascent throughout the circumpolar sea. The 

 heavily sampled shelf waters of the Bransfield Strait-North Graham Land area and the equally 

 heavily sampled South Georgia whaling grounds (apart from the November record) are barren of 

 these stages and so virtually also is the whole circumpolar sea east about from 30° W to 60° W. 

 In the East Wind drift the occurrence of a single First Calyptopis (a Norwegian record) is to be noted 

 off Enderby Land. 



January-February. It is evidently not until January (Fig. 74) that the Calyptopes in Weddell 

 West begin to reach the surface on a really substantial scale and by February it will be seen they are 

 found spread across this sector as far east as 30° W.^ Some now, both in January and February, 

 appear on the South Georgia whaling grounds, but in view of our failure to find either eggs or deep 

 recently hatched larvae in these island waters from November right through to February (Figs. 64, 

 65, 69 and 70) it seems certain that these could not have sprung from any local hatching but must have 

 got carried there in the surface stream. Apart from one substantial February occurrence in the slope 

 region of the East Wind drift off the Princess Martha Coast (near where as Fig. 70 shows the 

 eggs must also have been hatching) the surface waters of the circumpolar sea, whether shelf or oceanic, 

 are barren, or virtually barren, of Calyptopis forms. The negligible scale of their occurrences in the 

 heavily sampled shelf waters of the Bransfield Strait-North Graham Land and Ross Sea areas is again 

 particularly noticeable, further evidence (though of course from the same stations as in Fig. 70) 

 of the relative unimportance of the hatching (p. 201) that must take place in these shallow conditions. 



The concentration of the vast majority of the earliest surface arrivals in Weddell West, and their 

 virtual absence at this time from other parts of the circumpolar sea, again strengthens the evidence 

 that a major spawning takes place somewhere in the western reaches of the Weddell stream and at the 

 same time demonstrates conclusively that the November-December spawning there (p. 290 and 

 Fig. 64) could emphatically not have been taking place simultaneously, undetected all round 

 Antarctica, under the vast winter ice-sheet which still girdles the continental land. 



1 One of them in fact appears just outside the Weddell drift, but this may only be an appearance since the water boundaries 

 plotted in these charts can only be described as approximate and no doubt vary considerably from month to month and from 

 year to year. 



^ The two January plots of the highest order of magnitude are from Norwegian sources. They were recorded by 

 Ruud (1932) in ' Vikingen' on 22 and 28 January 1930. The locality in which they occurred was not examined by us that 

 year. 



I 



