THE OLDER STAGES 179 



there is every reason to believe (p. 184) that in so far at least as the krill are concerned the eggs 

 unquestionably sink. 



In December 1930, at Station 540, Fraser (1936, p. 18, Table III) records that 2496 eggs of £". superba 

 were taken at a depth of between 500 and 250 m., the vertical net in which they were captured having 

 fished to within approximately 10 m.^ of the sea bed in the shallow shelf water of the western Weddell 

 drift where it turns eastward away from the northern tip of Graham Land. At the same station 

 the 250-100 m., 100-50 m. and 50-0 m. vertical nets yielded 478, 52 and 4 eggs respectively. As he 

 remarks later (p. iii) the vertical distribution of the catch in this locality suggests a concentration of 

 eggs near the bottom, a concentration which, having regard to the very near approach to the sea bed 

 of the lowermost net at this station, leads one strongly to suspect, as Fraser himself suggests, that 

 spawning in this instance took place on the bottom itself, the occurrences of eggs at the higher levels 

 being possibly due to vertical mixing of the water column which in this region, as in all shelf seas, 

 is known he says to be particularly effective. Fraser in general is in favour of deep spawning, pointing 

 out, as already remarked, that in spite of intensive search, both inside and outside the pack-ice, the 

 observations covering every month of the protracted breeding season of this species, not a single 

 concentration of eggs has yet been recorded in the top 100 m. of the Antarctic surface layer. He 

 remarks, too, on the occasional occurrence, together with eggs, of fully adult males and gravid females 

 in deep hauls of the vertical nets, concluding with the guarded statement that since they should be 

 found deep down at the same time as eggs are taken it might be inferred that their presence at these 

 levels is connected with spawning. Such evidence, however, as he is able to provide in favour of deep 

 spawning from the vertical distribution of the eggs themselves, is not very satisfactory, since apart 

 from the rich haul at Station 540 the numbers he found elsewhere, like the vast majority of those 

 subsequently recorded by Fry and me, are negligible. ' It is remarkable ', he writes, ' that compara- 

 tively few eggs have been found in the samples analysed, and although Rustad's suggestion of circum- 

 scribed shoaling "and the consequent concentration of the eggs in relatively small areas" may be 

 partly the reason I do not think that it is completely satisfactory. E. superba must be reproduced in 

 immense numbers to hold the key position it has in the ecology of Antarctic life. The yield of the 

 70-cm. vertical nets, with the possible exception of those at Station 540, is surely not indicative of the 

 normal concentration'. 



In his concluding remarks on the purely bathymetric aspects of this problem Fraser (1936, 

 pp. 1 12-13), while pointing out that much more positive evidence was required before any definite 

 solution to it could be put forward, refers briefly to the possibility that the eggs may go through the 

 major part of their development at depths beyond the reach of our deepest vertical nets,^ suggesting 

 that the negligible numbers he found above 100 m. might represent 'the scattered product of dispersal 

 of a much greater mass situated in still deeper water '. It may be he suggests that the eggs when laid 

 usually sink below 1000 m., adding that the great abundance in the 500-250 m. net at Station 540 

 was probably due to the fact that this particular net struck a place where the eggs had become concen- 

 trated through having been laid in somewhat shallower water than usual. If, however, there was 

 nothing unusual in the discovery of this near-bottom coastal concentration, and we have no grounds 

 for supposing there was, then the obvious inference to be drawn from it is that spawning in certain 

 instances at least is a shallow water coastal phenomenon and that in such conditions it might even 

 take place on the bottom itself. 



Bargmann (1945) remarks on the scarcity of her stage 7 A, that is, gravid or near gravid, females in 

 the samples she examined, and since by far the greater part of her material came from the top 100 m. 



''- The sounding at Station 540 was 510 m. 



^ At the time when he wrote they had not, except in rare instances, been fished below 1000 m. 



19-2 



