352 



Fishery Bulletin 89(3). 1991 



indicate these were wedge-tailed shearwaters Puffinus 

 pacificus. 



Discussion 



As yellowfin tuna grow, they become involved in chang- 

 ing species associations, some of which are polyspecific 

 in nature, i.e., like the associations studied elsewhere 

 that seem situation-dependent, opportunistic, and often 

 casual. This is reflected in the changing pattern in the 

 association rates (p) from log-fish to school-fish to 

 porpoise-fish sets on tuna; it is this pattern, rather than 

 the actual rate values, that is of significance. The rates 

 themselves should be used cautiously as they are 

 derived from data affected by observer interest and 

 from samples that were not large. 



Foraging associates 



Logs, or other floating objects, seem to provide oppor- 

 tune sites for development of species associations. Not 

 only can logs attract prey and predators, but they can 

 also drift to convergences where species gather. The 

 behavior of predators feeding near flotsam is not 

 simple, however. If the arrival rate of prey should 

 decrease sufficiently, a passively feeding predator may 

 switch to active, wide-ranging foraging (see Gerritsen 

 1984). 



The very common association of sharks with tuna and 

 the strong decrease in this association from log- to 

 school- to porpoise-fish sets suggest that sharks most- 

 ly encounter tuna near flotsam. Then, like gray reef 

 sharks (carcharhinidae) that follow feeding carangid 

 fish for leftovers (Enewetak Is., pers. observ.), these 

 sharks might follow the yellowfin as both scavengers 

 and predators. Such behavior is probably of decreas- 

 ing advantage as the tuna grow and forage more widely 

 and at faster speeds. The tuna themselves, in some 

 situations, may be attracted to sharks, as they are to 

 whale sharks (Stretta and Slepoukha 1986). 



The overall association rate for most other species 

 with yellowfin tuna averaged in the 10% range, in- 

 cluding that of billfish, rays, turtles, and smaller tunas. 

 The marlins probably follow tuna both as parasitic 

 foragers and as predators; they share many prey 

 species with tunas and also eat tunas, especially the 

 smaller specimens (see Shomura and Williams 1975). 

 Unlike the sharks, however, these powerful fish appear 

 to have little difficulty keeping up with fast-moving 

 tuna; their association rates did not appear different 

 among set types. Manta rays and turtles probably 

 represent the opposite situation, where the tuna initiate 

 and maintain the associations, perhaps as an extension 



of their proclivity to investigate flotsam. This may be 

 why rays tended to be taken more often in the school- 

 fish sets. The large schools of black skipjack, bullet, and 

 skipjack tunas that are frequently taken with log- 

 associated yellowfin may be obtaining feeding and pro- 

 tective advantages, but these benefits likely decrease 

 as the yellowfin grow. The smallest tunas would find 

 it increasingly difficult to swim at the speed of the 

 yellowfin, and the danger of predation by the larger 

 fish would also increase. It is first the black and bullet 

 tunas that decrease in school-fish sets; finally even 

 skipjack become scarce in the sets on the large, 

 porpoise-associated yellowfin. 



The most conspicuous and strong association with 

 tuna is that of seabirds. Most bird species can feed in- 

 dependently of tuna, but they feed closely with these 

 fish at every opportunity. The flocks of terns and 

 smaller shearwaters that feed with free-swimming tuna 

 of school-fish sets suggest that these fish of mainly 

 nearshore waters feed on abundant, smaller prey. The 

 larger, more mature tuna are farther offshore, feeding 

 on sparser but evidently still-rich food patches. 

 Dolphins and the larger seabirds, i.e., boobies and 

 wedge-tailed shearwaters, feed with them (see also Au 

 and Pitman 1986). The dominating importance of 

 boobies in both porpoise-fish and log-fish sets suggests 

 they forage by a wide and fast-ranging search for all 

 discontinuities at the sea surface, including that of 

 surface-schooling tuna and dolphins. Such foraging 

 would be particularly effective offshore, for the 

 yellowfin there have been found to prey more on fish 

 (i.e., on larger but likely more patchily distributed prey) 

 and less on crustaceans than do nearshore-caught 

 yellowfin (Olson and Boggs 1986). 



The porpoise-fish sets are themselves a category of 

 species association, although not treated above as such. 

 The highly mobile dolphins and the tuna of these sets 

 appear to feed actively together; they share many of 

 the same prey species (Perrin et al. 1973). 



Enhanced foraging may be the main advantage of 

 polyspecific associations involving the larger, more 

 mobile species of tropical, pelagic seas; this may stem 

 from converging foraging tactics among these animals. 

 Just as boobies and shearwaters race about over a 

 feeding tuna school to maximize interceptions of 

 fleeting and unpredictably surfacing prey, yellowfin 

 and other large species, on a larger scale, should range 

 rapidly over large expanses to find sparse and un- 

 predictable, yet relatively rich, prey patches. Food 

 overlap would be increased by such nomadic foraging 

 (see Huey and Pianka 1981), and strong patchiness of 

 prey should itself reduce the tendency of each species 

 to exploit different spatial intervals of the spectrum 

 of prey distribution (see Terborgh and Stern 1987). 



