it is thought that i/i^en day brea'cs and the sea becomes faintly lighted 

 the schools suddenly begin to become active and seek foodj thereafter 

 until the time of n^ax their biting becomes better in direct proportion 

 as the light in the sea increases. Accordingly the catch increases sharply 

 and reaches its maximum at t^ s 6iOO - 8:00 a.m„^2) 



Thereafter, if we equate it with the passage of time, it falls off in 

 roughly a regression line„ It is thought that of A^ X2 ^3 and X^ 

 the main factor is Al followed by A3, and that in general X 2 and 

 A^^ 1^0 not show very marked values. Accordingly 



n = rimaxe" ( >i / A2 / X3 / \^)t '==, nmaxe" '^ I't . , (5) 



Actually, as set forth in (l), the fish school and feed most actively in 

 the early morning, and once their stomachs are filled their appetites 

 decline, but around noon, for reasons connected with the time required 

 for digestion, their appetites again increase and they become slightly 

 active. It is presumed that another such phenomenon is exhibited in the 

 evening just before sunseto Probably further study of what has been 

 simply represented as Ai will reveal that it must be represented as a 

 function shoiving three undulatory changes during the day. 



(3) Examination of the relationship between T and the catch n. As can be 

 seen in Figure 3, where the catch n is plotted against the time of duration 

 of fishing T, the greatest total number of fish are taken and the greatest 

 number of catches are made where T is one hour or less, with the maximum 

 where T is 10 - ^0 minutes. The n per time fished also increases in 

 direct proportion to the increase of T where T is 80 minutes or less. 

 Where T is greater than 80 minutes, however, neither the total number of 

 fish taken nor the n per time fished shovr any increase » If we consider 

 the catch for each 10 minutes of fishing, its value is very small where 

 T is greater than 80 minutes. It is not yet clear how this T is related 

 to such factors as the size of the school, and its swimiriing speed and 

 movements with regard to the boat. 



The correlation between the tide and the catch has been experienced 

 and remarked upon by many fishermen, being particularly emphasized v/here 

 fishing is carried on in the vicinity of islands, but I would like to 

 leave this question to a later time because no quantitative study of it 

 has as yet been completed. 



In conclusion I wish to express ray thanks to Mr. Takeo Maruyama for 

 his assistance with the calculations and drawings for this study. 



^2) Mr. Yasuo Suehiro has reportied that skipjack appear to bite somewhat 

 better in the early morning than during the dayo (Fisheries Experiment 

 Station Report No, 9, p. 99o 1938). 



U 



