1^8 DISCOVERY REPORTS 



They note, however, that the break-up of the winter ice is unpredictable. In 1956 it occurred 

 in early February, but in 1954 it did not break out until March. They note, too, that the water 

 did not remain unfrozen for long. In 1956, even before February was out, 'a skin of new 

 ice was already forming again and the rocks edging the sea were white with frozen spray and 

 tide- wash'. 



Depth of spawning 



In his study of the vertical net samples brought back by the 'Norvegia' expedition of 1928-9 

 Rustad (1930) found that the Nauplii and Metanauplii of Thysanoessa macrura G. O. Sars were con- 

 fined exclusively to depths below 200 m., the majority to the 400-200 m. layer which, with one 

 exception, was the deepest to be sampled by this expedition. In view of these findings he advanced 

 the hypothesis that T. macrura spawns below 400 m., the resultant larvae rising and eventually 

 reaching the surface when so far developed^ that they are able to feed on the phytoplankton. Owing 

 to the poverty of his material he was unable to come to any definite conclusion about the spawning of 

 E. superba. He seems, however, to have considered it unlikely that it would be a deep spawner from 

 his remark, ' It must however be kept in mind that E. superba with our present knowledge must be 

 regarded as a typical surface form'. Ruud (1932) rejected this hypothesis, objecting that the numbers 

 of deep Nauplii and Metanauplii Rustad found, 135 in all, were inadequate and, like the very small 

 numbers of krill Nauplii and Metanauplii he (Ruud) found at the same level, could only be regarded 

 as stragglers that had wandered away from a much larger population which, along with the eggs 

 he said, would be found, if a proper search were made for it, close to or right under the drifting 

 ice. In other words he substituted his own hypothesis that it would be there, directly under the 

 ice, that T. macrura, like E. superba, would spawn. In postulating this hypothesis Ruud seems 

 to have been much preoccupied with the assumption that the water directly under the ice is difficult 

 or impossible to explore. As Rustad (1934) quite properly points out, however, although there is 

 considerable truth in this assumption, it cannot always be upheld. For even if there zvere concentra- 

 tions of eggs hidden away beneath the floes they could hardly be expected to remain permanently 

 inaccessible there, since the floes must sometimes shift their position relative to that of the water, 

 in other words must sometimes travel through, as well as with, the water on which they float. 

 A great deal of exploration has in fact been done both close to and inside the pack, particularly by 

 the vessels of the former Discovery Committee, but so far, as Fraser (1936, p. 112) remarks, has 

 failed completely to reveal any indication of surface spawning by E. superba in this particular 

 environment. 



In his later paper Rustad (1934) extended his hypothesis regarding the deep spawning of T. macrura 

 to include E. superba. He writes, ' It seems, however, that the opposite supposition [i.e. deep as 

 opposed to Ruud's supposed surface spawning], viz. that E. superba and T. macrura spawn below the 

 Antarctic surface layer — is more easily brought in accordance with the known facts. In this case the 

 state of affairs will develop as follows : The developing eggs will ascend, and the major part will hatch, 

 e.g. between 400 and 200 metres, where the youngest larvae are accordingly to be found. The larvae 

 continue developing and ascending, and they finally reach the upper layers when they are able to 

 feed'. While he was undoubtedly correct in supposing the newly hatched larvae of both species to 

 rise, his nets, however, not going deep enough for him to visualise with any accuracy the actual 

 level from which the ascent would begin, Rustad's ascending eggs must be regarded as somewhat 

 hypothetical, for, as he admits himself, his material, a mere 16 eggs altogether, was far too scanty to 

 permit him to conclude that any such thing could take place. Indeed, as will be shown presently, 



1 That is, having reached the First Calyptopis stage. 



