MEPS 323:223-231 (2006)  -  doi:10.3354/meps323223

ABSTRACT: To test whether predation is an opportunistic size-based process within a tuna community, analyses were carried out on the size composition of stomach contents of bigeye tuna Thunnus obesus (Lowe, 1839) and yellowfin tuna T. albacares (Bonnaterre, 1788) caught in 1995 to 1997 during longline scientific surveys in the French Polynesian Exclusive Economic Zone (EEZ). Prey size distributions were compared with the size distribution of organisms collected by pelagic trawls carried out during the same programme. Relationships between prey size and predator size were studied using quantile regressions, and were related to tuna mouth-gape measurements. The results showed that mean and maximum sizes of prey increased with increasing predator size, and that maximum prey sizes (versus tuna size) were below those predicted by tuna mouth-gape size. Minimum prey size varied little with tuna size, and the size distributions of prey in tuna stomachs were very asymmetrical (lognormal type), confirming that during growth, tunas continue to feed on small prey. Comparison with previous studies on other piscivorous species from different ecosystems underlined that tunas feed on very small prey in relation to their own size. However, comparison of size distributions of prey in stomach contents and prey in pelagic trawls revealed that bigeye tuna select larger prey than yellowfin tuna when such prey are available.

KEY WORDS: Tuna · Thunnus spp. · Size-based predation · Diet · Quantile regression