Baja California. In the second model (passive migration), 

 he proposed that larval and juvenile skipjack tuna are 

 passively carried eastward in equatorial countercur- 

 rent(s), but he admits to problems in recruitment of fish 

 to the northern fishery from the terminus of the counter- 

 current. In the third model (gyral migration), developed 

 after personal communication with me when this report 

 was in the early stages of preparation, he proposed that 

 fish in the northeastern Pacific fishery move around a 

 zonally narrow counterclockwise equatorial gyre consist- 

 ing of the NECC and the NEC. In this model, he 

 considered that skipjack tuna spawn in northern spring 

 and summer in equatorial waters west of long. 130°W; 

 the larvae and early juveniles are carried eastward in 

 the NECC to the area south of the Revilla Gigedo 

 Islands where the juveniles, now of near recruitment 

 size, actively migrate out of the gyre into the Revilla 

 Gigedo waters and subsequently into the feeding 

 grounds off Baja California; and with the advance of 

 cold water in late fall the skipjack tuna leave the area 

 via the California Current Extension (CCE) to return to 

 the equatorial spawning grounds. He also proposed that 

 skipjack tuna forming the southern group spawn in the 

 southern spring and summer in the central equatorial 

 Pacific; that the larvae and juveniles north of the 

 equator enter the NECC and are carried eastward were, 

 on reaching the terminus of the NECC, the juveniles active- 

 ly migrate into the feeding areas off Central and South 

 America, leaving there via the SEC; and that larvae and 

 juveniles south of the equator are carried eastward in the 

 South Equatorial Countercurrent (SECC) and return to the 

 central Pacific as adults in the SEC. 



Kasahara (1968) cited the hypothetical movements of 

 skipjack tuna on a Pacific-wide basis developed by Naga- 

 numa (Unpubl. manuscr., see footnote 2). The move- 

 ments are based on catch as well as monthly variations 

 in length frequencies of skipjack tuna taken in the 

 Japanese longline fishery in 1965 (Fig. 11). 



Of the seven hypotheses above, only that by Naga- 

 numa (in Kasahara 1968) shows the migration of skip- 

 jack tuna on an oceanwide basis; others deal with more 

 restricted areas in the Pacific which contain today's 

 major fisheries or discuss movement in general without 

 indicating actual migratory routes. For the most part, 

 however, the routes proposed by Nagnuma in the west- 

 ern and central North Pacific and eastern South Pacific 

 are incomplete, due to his reliance upon only one year's 

 data, and his proposed route in the central South Pacific 

 could be opposite to the direction indicated (see follow- 

 ing discussion). 



Movement of Skipjack Tuna 

 Inferred from Present Study 



To detect movement, the positions of high (>4x 

 mean) CPUE were noted and connected by arrows in 

 quarterly sequence beginning with the earliest year 

 (Fig. 12), on the assumption that the quarterly changes 

 in the high -CPUE postions reflected actual movement of 

 fish. In some areas lacking in high-CPUE cells, above- 

 average (2 X -4 X mean) CPUEs were followed to complete 

 the movement pattern. In the figure, areas of above- 

 average CPUE in the first quarter of two or more years 

 were shaded to show that high-CPUE cells are not 



Figure 11. —Migration routes of skipjack tuna proposed by Naganuma (unpublished manuscript, see text footnote 2) based on 



tuna longline data. [Reproduced from Kasahara (1968).] 



21 



