Table 2. --Relative abundance between two groups 

 *: No data available 



A"b", but in the others no clear-cut basis for 

 making the division could be discerned so the 

 division was performed mechanically in the 

 following manner. For A'b', when the length 

 group at 101 to 104 cm. was taken at the center, 

 a point at which the curve of the line changed 

 was observed and this was taken as the bound- 

 ary of the groups. The length group of 101 to 

 104 cm. was split into two parts, which were 

 added respectively to A' *'^<i b'. In the case of 

 ajAi and a2A2, the length class standing nnid- 

 way between them was taken as the boundary 

 and the length group at that boundary was di- 

 vided into two, which were added respectively 

 to the group of large fish and the group of 

 smaller fish. According to the table, for all 

 sea areas totaled, in ajAi and a2A2 the pro- 

 portion of ai and a2 was less and the proportion 

 of Aj and A2 was greater. In contrast to this, 

 for A'b' and A"b", A' and A" were in greater 

 proportion and b' apd b" in the lesser propor- 

 tion. Comparing these results with those which 

 were obtained for each sea area separately, the 

 only example in which the two sets of results 

 were in disagreement was at ajAi for the area 

 of 165 E. to 180 . In all other cases the re- 

 lationships were found to be in agreement. To 

 sum up, in general for ajAj and a^z^Z *^^ 

 proportion of the group of smaller fish was less 

 and that of the group of larger fish was greater, 

 whereas for A'b' andA"b" the proportion of the 

 group of smaller fish was larger and the pro- 

 portion of the group of larger fish was smaller. 

 Where one would think of linking these relation- 

 ships directly to phenomena accompanying 

 growth, this phenomenon is clearly 

 incompatible. However, considered in the 

 following manner it will be realized that the 



aforedescribed relationship is not necessarily 

 contradictory. If we try comparing for sepa- 

 rate years the relative manner in which the 

 length groups appear which correspond respec- 

 tively to ai, a2 andAj, A2, we cansee through- 

 out the data for the last 5 years that in both 

 cases the former is small and the latter is 

 large. From the point of view of age, the 

 length groups corresponding to ai and a2 are 

 clearly younger than those corresponding to Ax 

 and A2, and the fact that the modes for the 

 younger groups of fish are always smaller than 

 those of the older groups of fish can probably 

 be ascribed to the problem of availability. 

 Specifically, it is postulated that the main rea- 

 sons are probably that the length groups corre- 

 sponding to ai and a2 come into the fishing 

 ground in lesser volume or that they are sub- 

 ject to the operation of too low a catching ef- 

 ficiency because of the selectivity of the gear. 

 It is not possible to offer a quantitative hypoth- 

 esis in regard to what the degree of the effect 

 of availability may be, but that it appears to be 

 rather well marked can be assumed by analogy 

 from the differences detected in the relative 

 manner of appearance of length groups corre- 

 sponding to 2l\ and a2 and Ai and A2i as seen in 

 figure 4. 



It is clear from the foregoing observations 

 that the establishment of the assumption 

 (aiAi ^A'b', a2A2 ^A"b") can be un- 

 derstood to involve no particular contradictions. 



Concerning problem (b): 



A difficulty lies in thef^ct that whereas A'b' 

 and A"b" are each composed of two different 



29 



