176 DISCOVERY REPORTS 



March, while other moderately high March values occur between 130° and 140° E. The position of 

 the ice-edge could also play a part, in so far as it could prevent ready access to good feeding in some 

 sectors and facilitate it in others. 



And so as Hart (1934) has said, this small herbivorous crustacean forms a link in 'one of the simplest 

 food chains possible, the building up of the vast body of the whale being only one stage removed from 

 the organic fixation of the radiant energy of the sun ' by the diatoms on which it feeds. 



Faecal pellets 

 In the vertical samples I examined from the Weddell and East Wind zones I often found masses 

 of green faecal pellets ' packed with diatom skeletons, principally Fragilaria ', their enormous abund- 

 ance in some samples suggesting strongly they were the recent excreta of densely packed shoals of 

 E. superba. They occurred in more or less equal abundance from the surface down to the deepest level 

 (1000 m.) examined from which it seems distinctly likely, the krill itself being massed principally at the 

 surface, they were sinking. Moore (1933) examined a number of these pellets from the Bransfield Strait 

 and concluded they were almost certainly those of E. superba, adding that they agreed ' in form with those 

 in the gut of the animal ' as well as with others he had already described (Moore, 1 93 1 ) from other euphau- 

 sians from the Clyde. He notes, however, that no pellets of this type were observed in our Discovery 

 bottom deposits and this he suggests is ' probably due to their quick breakdown as was the case in the 

 Clyde '. Many too (Sheard, 1953 ; Vinogradov, 1955), before they have sunk very far, must be devoured 

 by other animals. It seems unlikely, however, that they contribute directly to the food of deep-water 

 animals since they probably disintegrate (Bogorov, 1957) before reaching the levels at which they live. 



Similar pellets containing Fragilaria (dominant), Coscinodiscus, Rhizosolenia and Distephamis 

 speculum were found all over the Ross Sea shelf, some of the samples running to ' millions '. No doubt 

 they were the excreta of the other krill, E. crystallorophias, which swarms at the surface there, and from 

 their distribution and relative abundance from the surface downwards they too were evidently sinking. 

 Distephanus speculum, which is also eaten by E. superba, I found to be particularly abundant in these 

 high latitudes. 



Spawning and hatching 

 The act of spawning 



The fact that the deepest concentrations of larvae we strike are so often very largely, sometimes 

 indeed all, at the same stage of development (Tables 13 and 14) suggests that such concentrations may 

 spring from batches of eggs that all hatch simultaneously or within a very short space of time. This in 

 turn suggests that spawning itself may be a phenomenon involving the mass shedding of eggs by 

 swarms of gravid females (p. 250, Fig. 53) within an equally short space of time. Such phenomena 

 are not unknown. I quote from the last paragraph of Korringa (1957) on Lunar Periodicity. 



To analyse cases of periodicity in reproduction, the annual rhythm (the length of the breeding season), the monthly 

 rhythm (periodicity correlated with tidal sequence or moonlight cycle), and the daily rhythm (concentrating spawning 

 or swarming to a certain well-defined hour of the day, or to certain phases of the tidal cycle) must be clearly dif- 

 ferentiated. In extreme cases these three rhythms in combination may resuh in a complete concentration of repro- 

 ductive activities in the species concerned, so that an entire population may spawn simultaneously during one single 

 hour a year. 



In so far as the krill are concerned I would, of course, for Korringa's entire population substitute 

 entire swarm. 



In her classic work on the male and female reproductive system Bargmann (1937, p. 348) counted 

 over 1 1,000 eggs each from two gravid females. Even so, she adds, 'the process of laying is evidently 

 a rapid one '. 



