NOTE Hernandez-Garcia: Diet of Xiphias gladius 



409 



Discussion 



Commercial longline fishing generally lasts for 12- 

 14 hours and the fish caught can remain alive for 

 many hours (Boggs, 1992); thus digestion of stom- 

 ach contents is continuous. Moreover, regurgitation 

 regularly occurs (Tibbo et al., 1961). This may bias 

 values of stomach fullness. The period of study was 

 limited in each area and sample sizes were not large, 

 precluding statistical analysis. However, the diet of 

 swordfish appeared to show substantial variation 

 between areas. In neritic areas (zone A) swordfish 

 preyed upon pelagic and benthic species offish and 

 squids, whereas in oceanic areas (zone B) they preyed 

 mainly on squids (Table 2). This pattern was also 

 observed by Maksimov ( 1969). In addition, Carey and 

 Robison (1981) showed that in neritic areas sword- 

 fish have a different ecology and migration pattern 

 from that found in oceanic waters. In neritic areas, 

 swordfish are found near the bottom during the day- 

 light, feeding mainly on benthic species (i.e. Capros 

 aper and Merluccius merluccius in this study, Table 

 3); at night, they move offshore to feed actively on 

 squid and other migrating fauna concentrated near the 

 surface (i.e. Todarodes sagittatus in this study, Table 

 4). In oceanic waters, swordfish undergo a diel vertical 

 migration reaching about 600 m at noon and ascend- 

 ing to shallow waters at night; they can prey actively 

 on oegopsid squids at both ends of the migration (Carey 

 and Robison, 1981). The mean number of cephalopods 

 per stomach was 0.7 in zone A, similar to data reported 

 by Moreira ( 1990). In zone B (oceanic waters) the mean 

 was 1.8, similar to the value of 2. 1 calculated from data 

 obtained by Guerra et al. (1993). In zone C, the mean 

 number of cephalopds per stomach was 3. Bane (in 

 Fonteneau and Marcille, 1991 ) also showed that cepha- 

 lopod-prey are more important in oceanic waters than 

 inshore for yellowfin tuna, Thunnus albacares. 



From analysis of degradation, otoliths were 

 strongly eroded by acid; large, thick otoliths dissolved 

 more slowly than small, thin ones. Cephalopod beaks 

 and eye lenses and fish eye lenses were not affected 

 by acid. Fish were the item with the highest index of 

 numerical importance (Table 2) in zone A. However, 

 the value obtained must be treated with caution be- 

 cause the otoliths of Micromesistius poutassou may 

 accumulate owing to the structure of the stomach wall 

 and the time necessary to dissolve them (Fig. 3). The 

 otoliths of Scomber japonicus were not found free, a 

 feature that accords with the results of the analysis of 

 degradation. Therefore, if evaluation offish biomass is 

 based on the presence of otoliths, biomass may be over- 

 estimated for fish with large and dense otoliths (i.e. M. 

 poutassou ). Fish with smaller and weak otoliths may 

 be underestimated as was observed by both da Silva 



and Neilson (1985) and Jobling and Breiby (1986). 

 However, the rate of degradation of otoliths in this study 

 may be overestimated because the acidity was greater 

 than the value reported in general for fish stomach acids 

 (pH=2. 0—3.0) after the digestive process has begun. 

 Because cephalopod eye lenses and fish eye lenses are 

 not affected by acid, as noted by Clarke (1986b), the 

 number of lenses found in a stomach may be a means 

 of estimating the relative importance of cephalopods 

 and fish in the diet. However, the percentage of occur- 

 rence offish (counted from whole fish and free otoliths) 

 was similar to that obtained by Azevedo (1989) and 

 Moreira ( 1990) in areas nearest to zone A. In zone B, if 

 eye lenses offish and cephalopods are considered, both 

 of which experience a similar rate of digestion, the rela- 

 tive importance offish in the diet is less than what was 

 indicated by data from the otoliths. 



The importance (based on % number) of Stheno- 

 teuthis pteropus in zone C is not surprising given the 

 large abundance of this species at these latitudes 

 (Voss, 1966; Zuev et al., 1985). The preferred tem- 

 perature range for swordfish is from 25° to 29°C 

 (Ovchinnikov, 1970) which is nearly the same as that 

 for adult squid females (24°-30°C, Zuev et al., 1985). 

 These temperatures are almost constant at these 

 latitudes throughout the year. Around zone C, there 

 is a complex system of currents and frontal zones 

 where there is a high biomass (Blackburn, 1965; 

 Ovchinnikov, 1970): in addition, high swordfish 

 CPUE values have been obtained in these areas by 

 the Spanish fishing fleet (Mejuto et al., 1991) and 

 the Japanese longline fishery ( Palko et al. , 198 1 ). There- 

 fore, the availability of prey may be partially respon- 

 sible for the distribution of swordfish in these waters. 



Guerra et al. (1993) found that Sthenoteuthis 

 pteropus was the most common species in the sword- 

 fish diet in the Northeastern Atlantic; this species is 

 possibly one of the unidentified ommastrephid spe- 

 cies found by Maksimov (1969). Todarodes sagittatus 

 and Ommastrephes bartrami were also found. These 

 three species ascend to the upper epipelagic layer 

 and feed actively on myctophids and small squid at 

 night (Nigmatullin et al., 1977; author, unpubl. data). 

 During the day, S. pteropus descend to 350 or 400 m 

 (Moiseyev, 1988). These ommastrephids have a high 

 growth rate and their life span is no longer than 1.5 

 years (Arkhipkin and Mikheev, 1992; Rosenberg et 

 al., 1980; Ishii, 1977). Therefore, squid may be an effi- 

 cient means of energy transfer in oceanic food webs. 



Acknowledgments 



Thanks are given to F. J. Portela, Captain of the F/V 

 Manuela Cervera and to J. Gutierrez for providing 



