DISTRIBUTION AND GENERAL NOTES ON THE SPECIES 



28s 



50:- 



40^ 



202- 



\oZ- 



r 

 



10 



In practice samples of a hake population are modified by the catching power 

 of the trawl used. An ' idealized ' trawl, that captured all of the fish above a 

 certain size in its path and allowed all the others to escape, can be imagined to 

 catch samples showing a length-frequency curve of the same pattern as our 

 ' ideal ' sample, but with the point of origin at (say) 40-0 cm. instead of the 

 length of the newly-hatched larvae. Unfortunately such ideal conditions do not 

 exist, and the selective action of a sampling instrument such as a commercial 

 otter trawl of normal mesh retains a varying proportion of the smaller fish 

 below the size at which all are captured. The proportions of these smaller fish 

 retained varies in regular fashion according to their size, and follows yet another 

 ogive curve, the ' selection ogive ' of the trawl used. Selection ogives for trawls 

 of diflFerent mesh taking European hake are given by Hickling (1933, p. 71, 

 fig. 38). It will be seen that as the size of the mesh is increased and the 

 selection ogives become centred higher in the length scale the curves slope less 

 steeply. Theoretically it would be possible to obtain length-frequency samples 

 approximating to a normal distribution by increasing the size of the mesh. 



By covering an ordinary trawl with shrimp netting, Hickling has been able to 

 obtain samples which, when treated cumulatively over five annual surveys of 

 the grounds to the south and west of Ireland, show length frequencies approxi- 

 mating to the smoothed curve one would expect in ' ideal ' samples for hake 

 upwards of two years old. No practical gear could sample the younger length 

 classes simultaneously, for the minute fry are pelagic, and the yearlings do not 

 inhabit the same grounds as the older fish. Curves of percentage length fre- 

 quency of European hake, derived from Hickling (1933, table IXa), are shown 

 in Fig. 20, where they may be compared with the age distribution (a partial 

 'life curve') upon which that particular series of mean lengths was based. 

 These data of Hickling's also give us a valuable demonstration that the numbers 

 of the two sexes of European hake are nearly equal, the actual ratio from that 

 set of figures showing a slight preponderance of males. If samples such as can 

 be obtained with a commercial trawl are considered, it is found that females 

 preponderate to a considerable extent, owing to the higher escape ratio of the 



smaller males. 



Unfortunately, such data as we have for Patagonian hake are not directly 

 comparable with those for the European hake set out in Hickling's table IXa, 

 but they do compare very well with mean values of the figures set out in his 



series of measurements 

 on commercial trawlers, 

 of European hake from 

 areas 1-5 (Hickling, 1933, 

 tables XI a, XI b). These 

 areas cover that part 

 of the European hake's 

 geographical range most 

 nearly comparable to the 

 areas best sampled by the 

 ' William. Scoresby ' when 



\\. ^^ 



80 



90 100 110 



Fig. 19. Imaginary length frequency curves for an 'ideal sample' of hake. 

 Solid line: smoothed curve; pecked line: with small class intervals. 



