Table 2. --Local differences in size composition 

 of North Pacific albacore (October, 1949-51) 



Notes: (1) Fork length in cnn. "North of 38 N. " 

 may be taken to mean 38 N. -40 N. 



(2) Figures in columns are percentage of size 

 class in catch. 



(3) Column headed (M) is mean weight in kan 

 /JB.27 IbsJJ. 



(4) Length-weight relationship: 70 cm. , about 

 1. 9 kan ; 80 cm. , about 2, 8 kan; 90 cm. , 

 about 4. 1 kan; 100 cm. , about 5. 8 kan; 

 110 cm., about 7.4 kan. These length 

 classes coincide approximately with Aikawa's 

 age groups. 



As is clear from table 2, the greater part of the catch in this 

 region, regardless of locality, is in the 85-95 cm. size class. There 

 is clearly a tendency for smaller fish in the west and larger fish in the 

 east to mingle with this mean size group, but this trend is a continuous 

 one and shows no conspicuous gaps. However, since table 2 combines 

 the results of measurements in the years from 1949 to 1951, it also 

 includes local movements of the fishing grounds from year to year. 

 Consequently, it is not possible to discuss, on the basis of these results, 

 Uda's hypothesis of three groups of schools. The data are by no means 

 adequate, but table 3 shows the local differences in albacore sizes for 

 the month of October 1950 alone. 



In table 3 the local mingling of size classes is approximately as 

 shown in table 2; that is, it is continuous without any sharp discontinuities. 



From tables 2 and 3 the following observations can be made: 



1. If the sizes of albacore taiken at the same latitudes are com- 

 pared from east to west, the mean size group is 85-95 cm. , but toward 

 the east the proportion of larger fish nnixed in with this catch increases, 

 and west of 150°E. smaller fish increase. 



15 



