170 DISCOVERY REPORTS 



I have described the vertical distribution, and vertical movements, of the krill as I believe them 

 to be revealed from our wealth of bathymetric data. Professor Hardy, with whom I have been in 

 close contact, is inclined to stress the vertical migration rather more than I do and since his views 

 involve an understanding of the whole phenomenon of swarming, which has yet to be described, 

 I present and discuss them on pp. 268-78 at the end of the section dealing with the reaction of 

 these animals, both as individuals and swarms, to the intrusion of ship and nets. 



Absolute density 

 Although our townets (p. 266) cannot reveal it, some, if only approximate, estimate of the natural 

 density of the krill in the sea can be formed (i) from the apparent density in a patch, and (2) from the 

 number and approximate dimensions of surface patches observed in a given area. During her crossing 

 of the Weddell Sea patchy region described on p. 150 observers in R.R.S. 'William Scoresby' 

 recorded the approximate dimensions of 77 patches scattered about an area estimated to measure 

 about 18,000 by 1000 yards. Assuming that all the patches there were actually accounted for, the 

 density of the surface population could be expressed by the following equation, 



where v is the total volume of the 77 patches, obtained from the product of their area and thickness, 

 n the number of krill in a cubic yard, and V the total volume of the surface stratum in which the 

 patches were scattered about. 



The 77 patches covered a total area of approximately 13,010 square yards, their total volume, 

 assuming each to be i yard thick, being approximately 13,010 cubic yards. Assuming a value of 

 46,656 (that is, one to the cubic inch) for n and that the krill instead of being concentrated in patches 

 were distributed broadcast evenly throughout the whole i-yard thick surface stratum in which the 

 patches were disposed, then 



^ 13,010x46,61:6 , . 11- 1 



D = ^ ~ — ^ = 34 euphausians to the cubic yard. 



18,000x1000 J-r t- J 



This represents, in the immediate vicinity of the surface alone, some 140 million euphausians in a square 



nautical mile of patchiness, enough perhaps (p. 145) to provide more than fifty average-sized whales 



with a stomachful each. The estimate is conservative since some of the patches, especially towards the 



farthermost limit of vision,^ must inevitably have been missed. 



The weight of 8800 krill taken from a swarm on the southern edge of this patchy region (Station 



WS 558), drained of its formalin fixative and with the surface moisture removed by lightly drying 



between two layers of blotting paper, was 540 g., a measured sample (Table 36) showing that the 



swarm consisted of half-grown, approximately i-year old, euphausians, principally of the over 20 mm. 



or staple size. Assuming the same length frequency in all 77 swarms recorded, and taking the number 



of euphausians below every square yard of surface to be 34, then, expressed in terms of grams per 



square metre, 



540x34x1-196 2.coe/m2 

 8800 5 ^'' ' 



which works out at 8502 kg. per square nautical mile. 



Now, while it is true that in this small part of the Weddell drift there was a most exceptional 

 display of visible swarms on the surface, there is no reason to think that the actual density of the 



' Throughout the 9-mile crossing of the patchy region the limit of effective vision was put at 500 yards to port and 500 to 

 starboard. 



