148 



Fishery Bulletin 105(1) 



20° N 



(both lower and upper) were identified 

 by reference to features given by Clarke 

 (1986) and by comparison with material 

 held in our own reference collection. 



Well-preserved specimens of H. pelagica 

 were identified from the special arrange- 

 ment of luminous spots on the ventral side 

 of the mantle (Nesis, 1987). Beaks from 

 those specimens identified from the spots 

 (reference beaks) allowed us to identify 

 almost all ommastrephid beaks found in 

 fish samples as belonging to H. pelagica. 

 Importantly, wings of the lower beaks 

 darkened at a small size, thus precluding 

 misidentification with beaks of other om- 

 mastrephid species that darken at larger 

 sizes, e.g., Sthenoteuthis pteropus (Clarke, 

 1986). Lower rostral length (LRL) of beaks 

 was measured to 0.1 mm with a vernier 

 caliper and the allometric equations given 

 by Clarke (1986) were used to estimate 

 dorsal mantle length (ML) and whole wet 

 mass (M) from LRL. Specimens of H. pe- 

 lagica were assumed to be adults at 46 mm ML and 

 larger (Dunning and Brandt, 1985). 



Dietary data are presented by using two calculation 

 techniques, namely the frequency of occurrence and the 

 percentage by number of each prey type. Data were sta- 

 tistically analyzed by using SYSTAT 9 (SPSS, Chicago, 

 IL). Values given are means (±SD). 



Results 



Most (97%) of the fish caught on longlines belonged to 

 10 different species of large oceanic predatory fishes, 

 including longnose lancetfish {Alepisaurus ferox), four 

 scombrids (wahoo [Acanthocybium solandri], albacore 

 [Thunnus alalunga], yellowfin tuna [T. albacares], and 

 bigeye tuna [T. obesus]), swordfish (Xiphias gladius), 

 and four istiophorids (sailfish [Istiophorus albicans], 

 blue marlin [Makaira nigricans], and white marlin [Tet- 

 rapturus albidus], and longbill spearfish [T. pfluegeri]). 

 Most of the fish (93%) contained fresh remains in their 

 stomachs. Fish prey dominated the diet by mass (>50%) 

 of eight predator species (Table 1). Fish and cephalopod 

 items were almost equally important in the diet of white 

 marlin, whereas fish, cephalopods, and crustaceans were 

 the main food sources of albacore. 



Cephalopods amounted to slightly more than 50% 

 of the diet by mass in one fish species (white marlin) 

 only. They were an important prey class (>10%) in five 

 other species and were still a minor, but significant 

 (a5-10%), portion of the food of the four remaining 

 fishes (Table 1). Overall, cephalopods (both fresh and 

 accumulated items) were found in most of the individu- 

 als (76%) and a total of 2701 cephalopod beaks were 

 identified from the stomach of 105 fish. Hyaloteuthis 

 pelagica was by far the most important cephalopod prey 

 of the community of large predatory fishes, amounting 



• Cruise I 

 A Cruise I 



 Cruise I 



Figure 1 



Locations of longline sets carried out in the central Atlantic Ocean 

 during three cruises of the RV Shoyo Maru between October and 

 December 2000. 



to more than 50% of the total number of cephalopods 

 (up to 93%) in six species (Table 2). Indeed, it was found 

 to be the main cephalopod prey in all the fishes, except 

 in bigeye tuna and lancetfish where it ranked second 

 and third, respectively. Hyaloteuthis pelagica was much 

 more abundant in the diet of fish caught in tropical wa- 

 ters (cruises II and III) than in the diet of individuals 

 fished in temperate waters (cruise I) (?!=1937 and 15 

 beaks, respectively). 



Other important cephalopod prey (>10% by number) 

 included the small onychoteuthid squid Walvisteiithis 

 (= Onykia) rancureli in the diets of bigeye tuna and 

 albacore, and the pelagic octopuses Japetella diaphana 

 and common blanket octopus Tremoctopus violaceus in 

 those of yellowfin and bigeye tunas, respectively. Three 

 other ommastrephid squids were identified from fish 

 stomach contents; they were two rare prey species, the 

 Atlantic bird squid Ornithoteuthis antillarum and the 

 orangeback flying squid Sthenoteuthis pteropus, as well 

 as the bait, Argentine shortfin squid Illex argentinus. 



All fishes fed upon the same size range of H. pe- 

 lagica, including both juvenile and adult squids (Ta- 

 ble 3, Fig. 2), but overall they segregated by prey- 

 ing on squids of different sizes (ANOVA on LRL, 

 F,g 9071 = 16.36, P<0.0001). Post hoc Tukey multiple 

 comparison tests showed three groups of predators: 

 yellowfin tuna and sailfish fed on smaller squids (51 

 mm and 3.9 g on average), bigeye tuna, white marlin, 

 and longbill spearfish fed on larger individuals (60, 

 59, and 58 mm; 6.6, 6.2, and 5.7 g, respectively), and 

 albacore and blue marlin fed on squids of intermedi- 

 ate sizes (54 mm and 4.7-4.8 g). Accordingly, bigeye 

 tuna, white marlin, and longbill spearfish fed more on 

 adult squids (89%, 92%, and 93% of the total number 

 of//, pelagica, respectively) than did albacore and blue 

 marlin (85% and 77%) and yellowfin tuna and sailfish 

 (62%. and 73%) (Table 3). 



