15 



B. HELLAND-HANSEN 



at all seasons; in 1905 they rose, though without reaching the high values 

 attained in 1903 and 1906. However in these representations of the average 

 nxunbers we cannot enter into any details either for the seasons or for the various 

 smaller areas: since the material is neither large enough nor distributed in such 

 a way as to make our calculations reliable. We are therefore also unable to 



35 4d ^5 



LLrSCiTM IM CE.NTIME.TE.RS 



Fig. 4 



undertake a quantitative investigation of any value of the average numbers in 

 the various areas at different seasons, or to consider the question of monthly or 

 seasonal variations. Statistics of the catches by market-trawlers will here present 

 a more reliable picture of the variations in the haddock-supply (cf. D'arcy 

 Thompson). 



2) The average numbers of the different haddock sizes 



The whole of the 68733 haddocks have been measured, and their lengths 

 recorded in centimeters. The average numbers per trawl-hour for each centimeter- 

 size without regard to either time or area have also been calculated. The results 

 of these calculations are shown in fig. 4. The curve in this figure gives us 

 accordingly the simplest expression our measurements can procure of the 

 average size-composition of the haddocks in the North Sea. In what follows we 

 will consider the numbers of the different sizes and the ordinary size-groups as 

 shown by this curve. 



