REACTION OF KRILL TO SHIP AND NETS 263 



avenue of to all appearances clear water in the wake in which the stern nets caught nothing, or next 

 to nothing. These repeated daytime failures to sample the swarms by conventional methods led it 

 will be recalled (p. 152) to the novel idea of lateral towing from a forward boom, the experiment 

 being accompanied by remarkable results. Enormous daylight surface catches, hitherto unobtainable 

 by short-warped surface towing from the stern, were now obtained, catches vastly outnumbering even 

 the night catches of the conventional surface net. Particulars of some of these samplings are given 

 in Table 55 which shows the numbers captured, principally (p. 234, Fig. 44) of the normally 

 escaping over 20 mm. class, during the special daylight operations (p. 152) conducted at Station 

 WS 540. 



Table 55. Station WS 540. Particulars of daylight sampling by lateral net of five separate, 



but adjacent, surface swarms 

 Duration of haul 



H.^ 



WOODf 



Table 56. Average yield of the upper {100-0 m.) oblique stramin nets in daylight and darkness. 

 Night-time data in roman type, daytime data in italics 



Sum of average 



* See note below Table 51. 



Clearly in the boom net we have an instrument of enormous sampling power, and one it seems that 

 could be used with equal advantage for sampling the older stages of the surface population either by 

 day or by night. The largest single night catch of the staple whale food obtained over the stern on 

 the surface was 200,160, obtained not in the brief moments of lateral fishing but only after towing for 

 half an hour through one mile of water. Moreover, this relatively enormous catch is quite exceptional, 

 being in fact the only instance of a surface gathering of large krill exceeding 100,000 our stern-fished 

 surface nets provide. Without it the average stern catch for animals of this size (p. 259, Table 50) 

 is a mere 840. It must be remembered, too, having regard to the natural spacing of the swarms in the 

 sea (p. 148), that the boom net fished for the same distance would have the opportunity of striking 

 not one swarm only, but perhaps two, three or even more, and that if it sampled all with the same 

 success as at Station WS 540 its average yield would manifestly be vastly greater than this. 



The enormously larger catching power of the lateral as compared with that of the stern surface net, 

 whether the latter be used by night or by day, must it seems be due to the fact that the former samples 

 an undisturbed swarm. ^ Even so it is surprising that the oblique (loo-o m.) nets, approaching by night 

 from the dark, and by day from the poorly illuminated depths below, do not sample the sinking 

 swarms with equal effect, their average yield (Table 56), although substantial enough by ordinary 



1 See pp. 152 and 264. 



