Continued movement eastward in subsequent quarters 

 could find these fish in Hawaiian waters. Any such 

 move, however, can be made only by fish of the 

 central-eastern Pacific subpopulation (NW3), since Fujino 

 (Kitasato University, Iwate Prefecture 022-1, Japan, 

 pers. commn.) has indicated that no fish of the western 

 Pacific subpopulation have been sampled in the Hawaiian 

 fishery. 



The migration paths of deep-swimming skipjack tuna 

 of the western subpopulation in the southern hemisphere 

 (SWl and SW2) closely resemble those of the surface 

 fishery postulated by Fujino (1972), even to the possibil- 

 ity of the southern groups moving across the equatorial 

 current system, which could result in the mixing of fish 

 of both hemispheres. The area west of Fiji seems to be 

 the dividing line between the western and central-east- 

 ern skipjack tuna subpopulations in the South Pacific. 



In the eastern Pacific, the migration of skipjack tuna 

 (NEl) into the northern fishery, based on longline catch- 

 es, is similar to that of the surface fishery as postulated 

 by Rothschild (1965) and Williams (1972). Even the 



ing to the longline data, adult skipjack tuna are present 

 as far south as lat. 20°S and as far east as long. 

 90°-95°W in the second and third quarters. The off- 

 spring of these fish could easily enter the wake system 

 of the Galapoagos islands by simply moving along with 

 the current (Fig. 13). Once there the move into the 

 South American coastal waters can be made actively, as 

 demonstrated by tagging results (Fink and Bayliff 1970, 

 fig. 90). 



The return route of skipjack tuna from the eastern 

 Pacific northern fishery to the equatorial central Pacific 

 spawning grounds is seen more clearly in the longline 

 data than in the schematic presentation by Williams 

 (1972). It seems that the westward movement of fish 

 extends farther west, past long. 140°W, and this is corro- 

 borated by the recovery near Christmas Island and 

 Hawaii of seven fish tagged in the eastern Pacific 

 fishery (Table 4). These recaptures indicate that at least 

 part of the fish returning to equatorial waters from the 

 eastern Pacific migrate to Hawaii and the Line Islands, 

 while the remainder could return eastward in the NECC. 



Table 4. — Skipjack tuna tagged in the eastern Pacific and recaptured in the central 



Pacific (Seckel 1972). 



latter's "food bridge" theory linking the area of the 

 NECC east of long. 120°W to the Revilla Gigedo Islands 

 and Baja California is suggested in the progression of 

 longline catches (indicated by -l-'s) from the second 

 through the third quarters (Fig. 4). Williams also hypo- 

 thesized the possibility of a surface SECC as an addi- 

 tional mechanism of eastward transport of juvenile skip- 

 jack tuna into the eastern Pacific southern fishery. 

 Although a surface SECC has been identified as extend- 

 ing from the Solomon Islands to the coast of South 

 America along lat. 10°S (Reid 1961), its occurrence at 

 the surface in the area east of long. 120°- 140° W has not 

 been clearly defined. Wyrtki (1965) suggested that the 

 SECC exists there only in subsurface layers, whereas 

 Tsuchiya (1970, 1972) doubted the existence of zonal 

 continuity all the way across the ocean. Recently, Tsu- 

 chiya (1974) recognized an eastward surface current 

 (average speed of 7 cm/s) along lat. 10°S between long. 

 112° and 90°W in the southern summer. It should be 

 noted, however, that the current may be too weak and 

 too short to play a major role in supplying recruits into 

 the eastern Pacific southern fishery. 



The longline data discussed herein not only demon- 

 strate a possible avenue of southern hemisphere skipjack 

 tuna migration into the southern fishery, but show this 

 to be possible without the presence of a SECC. Accord- 



Much of the westward migration of skipjack tuna 

 from the eastern Pacific to Hawaii probably occurs in 

 the area covered by Seckel's (1972) drift hypothesis, i.e., 

 lat. 10°-20°N, long. 120°-160°W. Seckel gave 21-23 mo 

 as the average minimum duration of westward drift through 

 this area, and cited the possibility of even shorter 

 drift periods if the skipjack tuna were caught in cells of 

 high westerly flow of more than 25 cm/s (Seckel, 1972, fig. 

 17). Such a flow could take the skipjack tuna well over 

 3,900 km (2,100 nautical miles) in 6 mo. Both the average 

 minimum drift period and the accelerated drift period are 

 reflected in the recoveries in Hawaii of fish tagged in the 

 eastern Pacific. Of six such recoveries, four were of 21- 

 28-mo duration and two were of 9-mo duration (Table 4). The 

 duration of skipjack tuna migration from northern Baja Cali- 

 fornia to Hawaii suggested by the tuna longline data is 6 mo, 

 similar to that attainable by fish in the cells of high 

 westerly flow. 



That the movement of the skipjack tuna in the above 

 area involves more than just being transported by the 

 surface currents is evident from the seasonal shifting of 

 the northern boundary of distribution of the adults. The 

 boundary shifts northward in the third and fourth quar- 

 ters and southward in the first and second quarters (Fig. 

 8). More likely than not, the skipjack tuna in this region 



25 



