THE OLDER STAGES 167 



surface from daylight to dusk (Table 34, series (a)) are simply an expression of the lessening measure 

 of evasion that the older krill can achieve as the strength of the daylight wanes (see also p. 259, 

 Tables 50 and 54). The large to very large catches obtained on the surface as soon as full dark- 

 ness set in (Table 34, series (b)), obtained, it may be emphasised again, in the same v^^ater traversed 

 by the last sbc hauls made in the waning hours of daylight before the ship was put about, provide 

 convincing evidence of the extent of the evasion that must have taken place during the daylight, sunset 

 and twilight hours. Turning again to Table 33, it will be seen that it is only in full darkness, and 

 not, as might have been expected, by day, that there seems to have been some concentration of the 



I o 



40 



70 

 100 

 250 -• 



YOUNG SWARMS DAY 



500 1000 



13 

 3 



a 



6 



V 



I 10 



£ 40- 



g 70- 



100 



250 



OLD SWARMS DAY 

 500 1000 



13 

 3 



2 

 6 



YOUNG SWARMS NIGHT 

 500 _ 1000 

 ~l7 



OLD SWARMS NIGHT 

 5O0 _ 1000 

 "?} 



2 

 2 



1500 



10 

 40- 



70 

 100- 

 250 



TOTAL STOCK DAY 



500 1000 



13 



3 



2 

 6 



I 10 



S 70 

 100 

 250 



TOTAL STOCK NIGHT 



500 _ 1000 



7J 



2 

 2 



Fig. 18. Diumal vertical distribution of the 10 and 22 month old swarms sampled at St. 1835, the number of surface and 

 subsurface net hauls made in dayUght and darkness being shown as in Fig. 17. 



krill at depths immediately below the surface, ^ such concentration as was recorded, however, 

 decreasing rapidly below the 20 m. level and virtually disappearing it seems at about 50 m. The 

 complete subsurface void, however, encountered between 50 and 10 m. from sunset to dusk again 

 places the reliability of these night subsurface samplings in some considerable measure of doubt, 

 a critical assessment of their value again having to take account of the obvious difficulty of sending an 

 open net through a dense surface population to a deeper level and ensuring that in the darkness it 

 does not sample some of this population on the way. 



The diurnal vertical distribution of the first and second year krill taken at Station 1835, based on 

 the figures given in Tables 33 and 34, is shown diagrammatically in Fig. 18 in which the histo- 

 grams representing the average gatherings at subsurface levels are again, as in Fig. 17, based on 

 a grouping of depths of tow into selected broad horizons. Ahhough Fig. 18 would clearly have 

 carried more conviction had it been based on heavier daytime subsurface sampHng there are, never- 

 theless, enough day observations below 10 m. to postulate that the apparent scarcity of both young 

 and old broods at the surface by day was not due to their having been concentrated at this particular 

 station at some deeper level between 10 and 250 m. If they had been it is surprising that not one of 

 the eleven daytime subsurface nets, and notably not one of the six fished below 100 m., gives the 

 slightest indication of their presence. It might again of course be objected that since our day obser- 

 vations did not on this occasion go deeper than 250 m. it was somewhere below this level that the 



^ The large catch at 10 m. (p. 165, note i) having probably been obtained at the surface itself. 



