Length frequencies of blgeye tuna taken during the nine motherahlp expedition* are 

 given in table 8 and figure 10. As in the instance of yellowfin tuna, only the larger fish (50-145 

 pounds) are well represented, and the individual samples differ in their means (111-132 cm.) and 

 modal group or groups, though without any evident systematic order in respect to either time t 

 space. The longline-caught bigeye in this region are some 20 pounds heavier than the yellowfin. 

 A difference in the same direction and of greater magnitude was noted in the central Pacific 

 (Murphy and Shomura 1953). 



The bulk of the albacore catch were 86-112 cnn. in total length (30-65 pounds) (tabl* 

 9, fig. 10), which contrasts sharply with, for instance, catches of 65-90 cm. (10-35 pounds) o££ 

 San Pedro, California (Brock 1943). These small albacore taken off the Pacific coast of the 

 U. S. are found at the surface, whereas albacore have not been reported at the surface in 

 equatorial waters. Perhaps the large, adult fish tend to occupy cooler, deeper strata of water 

 in the tropics, waters comparable in temperature to those at the surface in more northern 

 latitudes. 



The skipjack taken during the mothership expeditions are represented in the two 

 length frequencies in table 10 and one in figure 10. The length distribution of the skipjack app«*ra 

 to represent what might be expected if the sampling were done from a complete population, on* 

 in which each size category was proportionally represented, and with gear selection precluding 

 the capture of the smallest individuals— . There is considerable increase in sise of the second 

 skipjack length frequency (table 10). The displacennent of approximately 4 cna. in about 3-1/2 

 months elapsed time may represent growth. 



The skipjack length distribution (fig. 10) provides a strong argument against 

 attributing the shape of the yellowfin, bigeye, and albacore length frequencies to gear selectioa 

 but rather indicates habitat selection by size; this is based on the fact that the upper end of the 

 skipjack distribution almost coincides with the lower end of the distributions of these species. 

 It might be argued that small yellowfin, bigeye, and albacore are not taken because they feed o» 

 a group of organisms different from those the larger individuals feed on. However, Reintjes 

 and King (1953) have shown that the foods taken are not related to the size of the yellowfin within 

 the limits of the fish sizes taken by trolling, livebait fishing, and longlinlng. These observatioa* 

 permit the tentative conclusion that there is considerable habitat selection, possibly both vertical 

 and horizontal, between small and large tuna, with the possible exception of the skipjack. The 

 large, mature yellowfin, bigeye and albacore tend to occupy sub-surface waters in the tropica 

 and snnall, immature fish tend to occupy the surface waters, resulting in longllne catches of 

 yellowfin, bigeye, and albacore being composed in the main of large fish. There naay also be a 

 considerable horizontal displacement between small and large yellowfin and bigeye tuna, and 

 there certainly is in the instance of albacore, small individuals almost never being taken in the 

 tropics in the central and western Pacific and never having been recorded at the surface there. 



RELATION OF YELLOWFIN CATCH RATES TO THE ENVIRONMENT 



The catch rates of yellowfin tuna experienced by the ntiothership expeditlona hare b««a 

 shown to vary significantly with time and with latitude. Presumably these changing rates reflaet 

 changing local abundance, and it seems reasonable to assume that these are a reflection of 

 changes in the environment, particularly in view of the lack of evidence for change In the abaolirte 

 magnitude of the population. Of necessity we confine the bulk of our discussion to the persisteatlf 

 greater abundance of yellowfin tuna in the immediate vicinity of the Equator, restricting our dla- 

 cussion of tinne variation to speculation on the causes of an apparent decline In abundance between 

 the pre- and post-World War II periods. 



2/ Longline gear utilizes relatively large hooka and baits designed to catch large tuna and marlin. 

 It would not be expected to be an efficient sampling tool for fish less than about 55 cm. In 

 length (about 8 pounds). 



19 



