© 2006 International Council for the Exploration of the Sea
Food and feeding ecology of Northeast Atlantic swordfish (Xiphias gladius) off the Bay of Biscay
a Centre de Recherche sur les Ecosystèmes Littoraux Anthropisés U.M.R. 6217, Université de La Rochelle, 17071 La Rochelle, France
b Centre de Recherche sur les Mammifères Marins, Université de La Rochelle 17071 La Rochelle, France
*Correspondence to V. Ridoux: Centre de Recherche sur les Ecosystèmes Littoraux Anthropisés, U.M.R. 6217, Université de La Rochelle, 17071 La Rochelle, France; tel: +33 546 50 02 91; fax: +33 546 50 02 94. e-mail: vridoux{at}univ-lr.fr.
As part of a larger project on the feeding ecology of large pelagic predators off the Bay of Biscay, this study analyses the diet of the swordfish, Xiphias gladius. Stomachs were collected from 86 swordfish. The diet was analysed in terms of prey occurrence, relative abundance, reconstituted mass, and size distribution. It consisted mainly of fish, 40.5% by mass (%M) and cephalopods, 59.3%M; crustaceans, 0.2%M, were considered secondary prey. When considering only the fresh fraction to allow for differential digestion rates, these figures were 77.3%M, 22.7%M, and trace amounts, respectively. Lanternfish, including Notoscopelus kroeyeri and Symbolophorus veranyi, were abundant, but paralepidids, Atlantic pomfret (Brama brama), and the squid Todarodes sagittatus, Ommastrephes bartramii, and Gonatus steenstrupi, dominated the diet by mass. The overall prey size range was 11–1420 mm, but sizes of 60–360 mm accounted for 80% of the distribution by number and of 140–760 mm for 80% of the distribution by mass. Intraspecifically, larger swordfish ate larger prey as a result of a change in species composition of the diet. The swordfish appears to show feeding plasticity both between different areas and between animals in the same area.
Keywords: Bay of Biscay, diet, feeding ecology, Northeast Atlantic, swordfish, Xiphias gladius
Received 30 December 2004; accepted 17 March 2006.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
The swordfish (Xiphias gladius) is a large solitary pelagic predator with a cosmopolitan distribution between 45°N and 45°S (Palko et al., 1981). In the Atlantic three stocks are known: a northern and a southern stock, and a Mediterranean one (Chow and Takeyama, 2000), but the precise geographic distribution of this highly migratory species remains unclear. Only spawning, which is in tropical western waters (ICCAT, 2003), is now well known. The species forages near the surface at night (0–90 m) and as deep as 650 m, with a maximum depth recorded at 900 m, by day (Carey and Robison, 1981; Matsumoto et al., 2003; Takahashi et al., 2003).
In the Northeast Atlantic, the swordfish is the target of the Spanish longline fishery and also a secondary, yet to be marketed, catch of several fisheries targeting juvenile albacore (Thunnus alalunga) as they migrate across the Northeast Atlantic (ICCAT, 2003). More generally, the Northeast Atlantic swordfish is part of the pelagic top predator community, which mainly comprises the swordfish itself, the albacore, the common (Delphinus delphis) and striped (Stenella coerulaeoalba) dolphins, the blue shark (Prionace glauca), and smaller fish including juvenile wreckfish (Polyprion americanus) and Atlantic pomfret (Brama brama). These species have different status with regards to fisheries: albacore and swordfish are target species of directed fisheries, blue sharks, wreckfish, and pomfret are secondary catches in the tuna fishery and are only partially marketed, whereas the two dolphins are protected. Proper management of such a composite set of species requires an understanding of the ecology of each species and their relationships within the community, notably in terms of utilization and partitioning of pelagic habitat and resources.
This paper is part of an ongoing programme that aims to investigate the diet of these large pelagic predators off the Bay of Biscay, in order to understand the role of each in the ecosystem and the way they partition food among the community. Here we analyse and quantify the diet of the swordfish off western Europe, and look at intraspecific variation.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Stomachs were collected from 86 swordfish caught during a research survey aimed at studying the impact of the French albacore driftnet fishery on small cetaceans (Goujon et al., 1993). Sampling took place between June and August 1993 off the Bay of Biscay between longitudes 38°N and 50°N and latitudes 9°W and 22°W (Figure 1). Swordfish were caught at night in driftnets 2.5 km long with a 20-m vertical spread. Body size ranged from 79 to 226 cm (lower jaw to fork length; LJFL). Of the 77 swordfish for which body length was documented, 21 were considered adults (>6 years old; LJFL >177 cm, consistent with ICCAT, 2000 and Arocha et al., 2003), 27 were juvenile (aged 2–6 years; LJFL 120–177 cm), and 29 were first-year animals (<2 years; LJFL <120 cm). Stomachs were dissected onboard, ligatured, and stored deep-frozen (–20°C) in polythene bags, awaiting analysis later.
|
Sample analysis
Sample analysis was aimed at describing the diet in terms of prey occurrence, relative abundance, reconstituted mass, and size distribution, and followed the general procedure now standard for marine top predators (Pierce and Boyle, 1991; Croxall, 1993; Ridoux, 1994). Each stomach was weighed and its contents emptied into a tray, then its wall was weighed again to obtain the mass of the contents by difference. The state of repletion of the stomach was visually assessed on a four-level scale: full, >50% filled, <50% filled, empty. The stomach content was washed through a sieve of 0.2-mm mesh. The diagnostic parts, defined here as hard parts that provided clues for specific identification and body length/mass calculation, were recovered and stored dry in the case of fish bones and otoliths, or in 70% ethanol in the instance of cephalopod beaks, crustacean remains, and any remains with flesh attached. The items found were identified to the lowest possible taxon through the use of published guides (Lagardère, 1971; Clarke, 1986; Härkönen, 1986) and a reference collection of specimens obtained from commercial and scientific trawling in the Bay of Biscay and adjacent areas of the Atlantic Ocean. To minimize overestimating prey items resistant to digestion (e.g. cephalopod beaks: Bigg and Fawcett, 1985), each item was scored on a scale specific to the main prey type (fish, cephalopods, crustaceans), according to their state of decomposition. This allowed us to determine a fresh fraction that provided a better representation of the composition of the ingested prey than did the total stomach content.
The total number of food items was estimated as the greatest number given either by paired structures (otoliths, operculum, hyomandibular, premaxillary for fish, eyes for crustaceans) or by non-paired structures (parasphenoïd for fish, upper and lower beaks and gladii for cephalopods, carapace and telson for crustaceans). Diagnostic hard parts such as beaks, otoliths, and carapaces were measured with a digital vernier calliper (±0.02 mm) using accepted standards (otolith length and lower rostral length in mm, carapace length without rostrum in mm; Clarke, 1986; Härkönen, 1986). A random subsample of up to 30 diagnostic hard parts per prey species per stomach was measured.
Data analysis
The occurrence of a prey taxon was the number of stomachs observed in it. The relative abundance was given by the number of individuals of the same taxon found throughout the sample. The reconstituted mass was the product of the average reconstituted body mass and the number of individuals of the same taxon in each sample, summed throughout the sample set (Equations (1)–(3)).
|
| (1) |
|
| (2) |
|
| (3) |
ij is the average individual body mass of prey i in sample j, calculated from individual body masses of up to 30 individuals per sample. Individual prey body length and body mass were calculated using allometric relationships from the literature (Clarke, 1986; Härkönen, 1986) or were matched to measurements performed on specimens in our reference collection (University of La Rochelle, unpublished data). Standard length was used for fish, and dorsal mantle length (DML) for cephalopods when describing the general diet (Table 1), but total body length including arm length was derived from relationships between total length and mantle length in published illustrations of the corresponding species (Nesis, 1987) to describe overall prey size distribution. The latter qualifies the cephalopod prey size targeted by a predator (Figure 2) better than DML.
|
|
When the size of diagnostic hard parts in stomach contents was well below the lower limit of the size range given in published relationships, we assumed that the growth was isometric between the origin (diagnostic part size = 0; body length = 0; body mass = 0) and the lower limit of the range. In this way, aberrant results such as negative body lengths were avoided. Prey size distributions were formulated as both percentage by number and percentage by mass contributed by each size class, because the two variables convey quite different information about the importance to the diet of prey with varying body length.
A discriminant analysis was computed using an XLSTAT package to test whether and how swordfish life stages might be discriminated on the basis of their diet composition. Discriminant analysis generates linear combination (the so-called discriminant functions) of a set of explanatory variables (here the proportion by mass of each prey) that maximizes the discrimination between pre-defined groups (here the life stages; Quinn and Keough, 2002). The results may be presented graphically, one graph showing the correlation between explanatory variables and the discriminant functions, and a second graph the individuals in the discriminant function space.
Confidence intervals (95% CI in Table 1) around the percentages by number and by mass were generated for each prey species using bootstrap simulations (Reynolds and Aebischer, 1991). The bootstrap routine was written in R software (Ihaka and Gentleman, 1996). Random samples were drawn with replacement, and the procedure was repeated 300 times.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Overall, the degree of repletion was fairly high: 22 stomachs (25.6%) were full, 61 (70.9%) were about half-filled, and just 3 stomachs (3.5%) were empty. Subsequent analysis was conducted on the 83 non-empty stomachs. The mass of food remains totalled 46 780 g or 564 ± 222 g per sample. In general, food remains were fairly fresh, mostly prey remains with some flesh attached to hard parts.
Diet composition
In all, 1125 prey items were identified, accounting for a total reconstituted biomass of 152 966 g (Table 1). The diet was highly diverse, with 17 species of fish, 16 species of cephalopods, and 4 crustaceans.
Fish were found in 95.2% of the stomachs, representing 49.4% of the diet by number and 40.5% by mass (Table 1). Among them, the two myctophid species, Symbolophorus veranyi and Notoscopelus kroeyeri, accounted for fairly high proportions of the diet by number (13.1% and 9.3%, respectively), but just 1.0% and 0.9% by reconstituted mass, respectively (they are relatively small fish). In contrast, the barracudina (Paralepis atlantica) and the Atlantic pomfret (Brama brama) represented, respectively, only 4.2% and 2.8% by number, but 7.2% and 14.0% by mass.
Cephalopods were found in 75.9% of non-empty samples (Table 1); their contribution to the diet was 43.3% by number and 59.3% by reconstituted mass. There were mainly three species; Gonatus steenstrupi, Ommastrephes bartramii, and Todarodes sagittatus, accounted, respectively, for 28.9%, 1.3%, and 4.4% by number and 8.5%, 21.8%, and 19.6% by mass.
The main crustaceans were the euphausiid Meganyctiphanes norvegica and the pelagic shrimp Funchalia woodwardi (Table 1). Crustaceans were always found associated with fish and were also observed in fish prey gut contents; as a consequence they were considered as mostly likely fish prey secondarily eaten by swordfish and, therefore, they were not included in further analysis.
The composition of the fresh fraction differed from that of the total diet (Table 1). Indeed, the relationship between fish and cephalopods, both in terms of relative abundance and reconstituted biomass, was fairly balanced in the total diet, whereas fish dominated the fresh fraction. The cephalopod share dropped from 43.3% by number in the total diet to just 4.0% in the fresh fraction, and from 59.3% to 22.7% by mass. This change was matched by a corresponding increase in fish proportion, which reached 86.2% by number and 77.3% by mass in the fresh fraction. Major prey species of the fresh fraction were the paralepidids (24.1% by mass) and Atlantic pomfret (28.0%) among fish, and the ommastrephid O. bartramii (13.5% by mass) and T. sagittatus (7.4%) among squid.
Length distribution of prey
The overall prey size range was 11–1420 mm (Figure 2). The upper limit was a single specimen of the family Trachipteridae, measuring 1420 mm, which was found in a 181-cm LJFL swordfish. Within the overall range, 80% of the distribution by number measured 60–360 mm, with a conspicuous mode at 60–200 mm. The distribution by mass was broader, with 80% of the range between 140 and 760 mm, and there was no mode.
At an individual prey species level (Table 1, Figure 3), myctophids were within the smaller mode of the overall prey size distribution: S. veranyi ranged from 56 to 120 mm (87.4 ± 14.8 mm on average) and N. kroeyeri from 39.9 to 140.6 mm with a single mode at 90–120 mm (114.5 ± 26.2 mm). In contrast, the barracudina and the Atlantic pomfret were in the larger modes of the overall prey size distribution, P. atlantica being distributed between 252 and 530 mm (379.6 ± 57.3 mm) and B. brama from 220 to 440 mm (301.2 ± 47.8 mm). Of the squid, the three main species had very different body size distributions. G. steenstrupi ranged from 32.3 to 145.0 mm DML with a peak at 30–90 mm (77.7 ± 15.4 mm), T. sagittatus from 132.6 to 485.0 mm (265.2 ± 90.6 mm on average), and O. bartramii from 279.9 to 397.3 mm (322.3 ± 32.0 mm).
|
Variability in the diet
Larger swordfish tended to eat larger prey than small swordfish (Figure 4). Adults fed consistently on bigger prey species than juveniles (Mann–Whitney unilateral test: Z = 7.92, p < 0.0001). This shift in prey size mostly reflected a change in diet composition between first-year, juvenile, and adult swordfish (S. saurus and myctophids were eaten by smaller swordfish, and B. brama, ommastrephids, and garfish, Belone belone, by larger ones; Figure 5) rather than a selection of larger body size of prey within the same array of prey species.
|
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This study was the first quantitative analysis of swordfish diet off the Bay of Biscay. The overall diet consisted mainly of fish and cephalopods which contributed 40.5% and 59.3%, respectively, by reconstituted mass. However, in the fresh fraction, fish clearly dominated the diet with 86.2% by number and 77.3% by reconstituted biomass. The major prey species belonged to the mesopelagic diurnally migrating fauna of the northeastern Atlantic, such as myctophids (Whitehead et al., 1989) and ommastrephids (Nesis, 1987), in addition to some large epipelagic species such as Atlantic pomfret and some mesopelagic species, such as large paralepidids (Whitehead et al., 1989). The size range of prey eaten by swordfish was from 11 to 1420 mm, with size classes 60–200 mm the most abundant, but size classes 140–760 mm provided 80% of the reconstituted prey mass. Very large fish of the family Trachipteridae at body sizes of up to 2.5 m standard body length had already been reported from swordfish stomachs (Maksimov, 1968).
Samples were collected over a period of three months in a single year and were limited by the schedule of the driftnet fishery, which operated only at night. Therefore, neither interannual nor seasonal variations of the diet could be researched. Several other limitations are inherent to studying the diet of marine top predators by analysing stomach contents and have been described in detail by different authors (Bigg and Fawcett, 1985; Jobling and Breiby, 1986; Santos et al., 2001). These limitations mostly consist of differential gut transit time and digestibility among prey species. In the present work, it is expected that the fresh fraction would be less biased in its composition than the total stomach contents, so producing a more accurate picture of the feeding habits of the swordfish. For example, cephalopods were overestimated in the total content compared with in the fresh fraction, because they were found mostly as digested loose beaks. In contrast, other important prey species such as myctophids and paralepidids did not show much difference in their mass contribution to the fresh and total stomach contents because they were found in all stages of digestion, from quite fresh fish to fully digested remains. However, in our results, the composition of both the total diet and the fresh fraction was given, first to allow comparison with the results of other authors who did not separate prey remains on the basis of their state of digestion and, second, to provide theoretically less biased results for further studies requiring food composition data for modelling purposes.
Geographic variations of the diet
The data generally agreed with previous results, but the details show that extensive differences among studies arose as a result of different sampling habitats (Table 2). In Azorean waters, Clarke et al. (1995) showed a clear dominance of fish, but the main species, boarfish (Capros aper), was an epibenthic species associated with island-influenced waters. Near the Straits of Gibraltar, Hernandez-Garcia (1995) examined samples collected over shelf habitats and found that fish dominated the diet of swordfish, with 93.3% by number, and that neritic families such as gadids and merlucciids were more important than mesopelagic ones. In the Mediterranean Sea (Orsi Relini et al., 1994), both fish and cephalopods were the main components of swordfish diet, with a predominance of epi- and mesopelagic prey such as scomberesocids and myctophids. In the Northwest Atlantic (Stillwell and Kohler, 1985), cephalopods were by far the most important prey item, twice as abundant as found here. Crustaceans accounted for a limited part of the swordfish diet in all regions, except in the Mediterranean Sea (Orsi Relini et al., 1994), where the most important crustacean species in terms of mass was the penaeid shrimp Funchalia woodwardi. In all regions, the diversity of prey was fairly large, especially in the North Atlantic (18 fish and 21 cephalopod species around the Azores; 20 fish and 12 cephalopods in the Northwest Atlantic; 17 fish and 16 cephalopods in the present work), and less so in the Mediterranean Sea (10 fish and 9 cephalopod species in the Ligurian Sea; 7 fish and 13 cephalopods in the Straits of Gibraltar). References for these results are given in Table 2. In addition to regional variability, which reflects prey availability, this high prey diversity suggests a high individual plasticity in foraging behaviour.
|
Intraspecific variation in swordfish diet
The present work showed a significant change of dietary composition and prey size range among swordfish of increasing age and size. A positive correlation was found between predator and prey length (Figure 4), agreeing with the differences in food composition between juveniles and adults (Figure 5). Indeed, juveniles feed less on large prey such as Brama brama, Paralepis atlantica, and Todarodes sagittatus, characterizing a different niche from adults. This finding is different from previous conclusions. Indeed, Stillwell and Kohler (1985) showed no indication of food preference with respect to swordfish sex or size throughout their samples (the contents of 151 stomachs), all their sizes of swordfish feeding on various schooling and vertically migrating species such as lanternfish and baracudinas. Similarly, Clarke et al. (1995) found no difference in diet attributable to swordfish size.
Significance in terms of swordfish foraging strategy
Swordfish forage near the surface by night (0–90 m) and as deep as 650 m by day (Carey and Robison, 1981; Matsumoto et al., 2003). More recently, a swordfish recaptured with an archival tag was shown by temperature records to have dived to 900 m (Takahashi et al., 2003). Accordingly, the present study also suggests that feeding is mostly concentrated in epi- and mesopelagic waters. Indeed, among the main prey are species such as myctophid fish and ommastrephid squid, which are mesopelagic organisms that undertake daily migrations that bring them to the surface at night (Nesis, 1987; Whitehead et al., 1989), paralepidids, that are found in the mesopelagic area, and species restricted to epipelagic waters, such as Atlantic pomfret, garfish, and Atlantic saury.
The state of digestion of the prey suggests that both nocturnal and diurnal feeding were important and that, as expected, the foraging depth would be greater by day than by night. Clarke et al. (1995) suggest that swordfish in Azorean waters would forage at the surface by night but be inactive at depth by day. Considering that swordfish can reach depths where the mesopelagic fauna concentrates during daylight, it is possible that they take advantage of their main prey being more densely distributed and less active to catch them at a higher rate per unit of effort than when the same prey are actively looking for their own food at night in surface layers. Mesopelagic fish density 450 m deep by day is about four times greater than in the upper 100 m layer at dawn and dusk, the two periods of maximum fish density in the surface layer (Roe and Badcock, 1984). Both nocturnal and diurnal foraging by swordfish would be possible, and feeding decisions would be based on foraging success. Only a refined understanding of the day and night prey availability and catchability, notably using acoustic survey methodology, and of swordfish foraging behaviour and energetics could resolve this issue.
Another aspect of the swordfish foraging ecology relates to prey size, prey aggregation, and communal or solitary foraging. Foraging on prey species of small individual body mass is worthwhile only if the catch rate per unit of effort is sufficiently high; in many pelagic top predators this is achieved by communal hunting, which reduces the escape capability of prey. Solitary predators would tend to rely on prey of larger body mass. The importance of large prey in its diet (B. brama, P. atlantica, B. belone, trachipterid fish, ommastrephid squid) is consistent with a solitary foraging strategy (Palko et al., 1981).
Food partitioning with the albacore tuna
The diet of the swordfish documented here can be compared with that of another large fish predator, the albacore (Pusineri et al., 2005). Both species live sympatrically in this part of the ocean, and the samples were collected simultaneously and in the same area with the same equipment. The composition of the diet of both predators differs considerably in terms of species composition and prey size. Indeed, albacore feed mainly on small to medium-size pelagic prey (ranging between 6 and 228 mm, against 11–1420 mm for the swordfish) including Maurolicus muelleri and Scomberesox saurus, but excluding lanternfish; crustaceans are also an important part of the diet although it is questionable whether they are primary or secondary prey. Contrary to the swordfish, cephalopods are absent from the fresh part of albacore diet, so foraging on cephalopods is thought to be occasional. Specific diversity of the two diets also differs largely. While as many as 36 prey species were found in the swordfish diet, only 12 species were recorded for albacore. It could be argued that the immature albacore sampled ranged in size from 53 to 93 cm body length (Pusineri et al., 2005), whereas immature swordfish were 79–226 cm LJFL, and that this might account for the observed differences in dietary composition; however, even when only large albacore are compared with small swordfish, the conclusions are the same.
According to these comparisons, the swordfish appears to live at a generally higher trophic level than albacore, as suggested by the larger size of its prey. Accordingly, Hernandez-Garcia (1995) reported tunas (Thunnus sp.) as occasional prey of the swordfish. The two species can reach fairly different depths: immature albacore are physiologically constrained to living at depths shallower than 100 m (Aloncle and Delaporte, 1973; Alonso et al., 2005), whereas the swordfish can feed as deep as 650–900 m (Carey and Robison, 1981; Matsumoto et al., 2003; Takahashi et al., 2003). Finally, this is in line with the foraging tactics of both species; juvenile albacore live and forage in large aggregations (Bard, 1981), whereas swordfish forage solitarily (Palko et al., 1981).
|
Acknowledgements |
|---|
The work formed part of a large research programme on the role of pelagic top predators in the Bay of Biscay and adjacent Atlantic Ocean. Funding from IFREMER and CNRS was obtained from the research project Chantier Golfe de Gascogne, Programme National d'Environnement Côtier. Identification of prey reference specimens was checked by Jean-Paul Lagardère (CREMA, L'Houmeau, France), Jean-Claude Quéro (Muséum d'Histoire Naturelle, La Rochelle, France), Yves Cherel (CNRS, Chizé, France), and Begoña Santos Vasquez (University of Aberdeen, Aberdeen, UK). Grégoire Certain helped with the bootstrap analyses. All offered their support, their knowledge, and their time, and are gratefully acknowledged for their contribution.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Aloncle H. and Delaporte F. (1973) Rythmes alimentaires et circadiens chez le germon Thunnus alalunga dans le Nord-Est atlantique. PhD thesis, Paris VI.
Alonso C., Arrizabalaga H., Restrepo V.R. (2005) Contribution of a chapter on albacore tuna for the revised ICCAT field manual. Collective Volume of Scientific Papers International Commission for the Conservation of Atlantic Tunas 58:1646–1669.
Arocha F., Moreno C., Beerkircher L., Lee D.W., Marcano L. (2003) Update on growth estimates on swordfish, Xiphias gladius, in the Northwestern Atlantic. Collective Volume of Scientific Papers International Commission for the Conservation of Atlantic Tunas 55:1416–1429.
Bard F. X. (1981) Le thon germon (Thunnus alalunga) de l'Océan Atlantique. Doctorate thesis in biology, Paris VI. 329 pp.
Bigg M.A. and Fawcett I. (1985) Two biases in diet determination of northern fur seals. In Beddington J.R., Beverton R.J.H., Lavigne D.M. (Eds.). Marine Mammals and Fisheries(George Allen and Unwin, Boston) pp. 277–282.
Carey F.G. and Robison B.H. (1981) Daily patterns in the activities of swordfish, Xiphias gladius, observed by acoustic telemetry. Fishery Bulletin US 79:277–292.
Chow S. and Takeyama H. (2000) Nuclear and mitochondrial DNA analyses reveal four genetically separated breeding units of the swordfish. Journal of Fish Biology 56:1087–1098.[CrossRef][Web of Science]
Clarke M.R. (1986) A Handbook for the Identification of Cephalopod Beaks(Clarendon Press, Oxford) 274 pp.
Clarke M.R., Clarke D.C., Martins H.R., Da Silva H.M. (1995) The diet of the swordfish (Xiphias gladius) in Azorean waters. Arquipélago. Life and Marine Sciences 13:A53–69.
Croxall J.P. (1993) Diet. In Laws R.M. (Ed.). Antarctic SealsCambridge University Press pp. 268–290.
Goujon M., Antoine L., Collet A., Fifas S. (1993) Approche de l'impact écologique de la pêche thonière au filet maillant dérivant en Atlantique nord-est. Ifremer, Brest, Rapport interne de la Direction des Ressources Vivantes, RI.DRV-93.034-RH/Brest. 47 pp.
Härkönen T.J. (1986) Guide to the Otoliths of the Bony Fishes of the Northeast Atlantic(Danbiu ApS, Hellerupe) 256 pp.
Hernandez-Garcia V. (1995) The diet of the swordfish Xiphias gladius Linnaeus, 1758, in the central east Atlantic, with emphasis on the role of cephalopods. Fishery Bulletin US 93:403–411.
ICCAT. (2000) Detailed report on swordfish. Report of the ICCAT swordfish stock assessment session (Madrid, Spain, September 27–October 4, 1999). Collective Volume of Scientific Papers International Commission for the Conservation of Atlantic Tunas 51:1001–1209.
ICCAT. (2003) Report of the 2002 Atlantic Swordfish stock assessment session. Collective Volume of Scientific Papers International Commission for the Conservation of Atlantic Tunas 54:1289–1415.
Ihaka R. and Gentleman R. (1996) R: a language for data analysis and graphics. Journal of Computational Statistics and Graphical Analysis 5:299–314.[CrossRef]
Jobling M. and Breiby A. (1986) The use and abuse of fish otoliths in studies of feeding habits of marine piscivores. Sarsia 71:265–274.[Web of Science]
Lagardère J.P. (1971) Les crevettes des côtes du Maroc. Travaux de l'Institut Scientifique Chérifien et de la Faculté des Sciences, Séries Zoologiques 36:1–140.
Maksimov V.P. (1968) Swordfish attack on a shark. Problems of Ichthyology 8:756.
Matsumoto T., Saito H., Miyabe N. (2003) Report of observer program for Japanese tuna longline fishery in the Atlantic Ocean from September 2001 to March 2003. Collective Volume of Scientific Papers International Commission for the Conservation of Atlantic Tunas 55:1679–1718.
Nesis K.N. (1987) In Levitov B.S. (Ed.). Cephalopods of the World: Squids, Cuttlefishes, Octopus, and Allies(TFH Publications, Neptune City, NJ/Moscow).
Orsi Relini L., Garibaldi F., Cima C., Palandri G. (1994) Feeding of the swordfish, the bluefin and other pelagic nekton in the western Ligurian sea. Collective Volume of Scientific Papers International Commission for the Conservation of Atlantic Tunas 44:283–286.
Palko B.J., Beardsley G.L., Richard W.J. (1981) Synopsis of the biology of the swordfish Xiphias gladius Linnaeus. NOAA Technical Report NMFS circular 441: 21 pp.
Pierce G.J. and Boyle P.R. (1991) A review of methods for diet analysis in piscivorous marine mammals. Oceanography and Marine Biology 29:409–486.
Pusineri C., Vasseur Y., Hassani S., Meynier L., Spitz J., Ridoux V. (2005) The food and feeding ecology of juvenile northeast Atlantic albacore tuna, Thunnus alalunga, off the Bay of Biscay. ICES Journal of Marine Science 62:116–122.
Quinn G.P. and Keough M.J. (2002) Experimental Design and Data Analysis for BiologistsCambridge University Press 537 pp.
Reynolds J.C. and Aebischer N.J. (1991) Comparison and quantification of carnivore diet by faecal analysis: a critique, with recommendations, based on a study of the fox Vulpes vulpes. Mammal Review 21:97–122.[Web of Science]
Ridoux V. (1994) The diets and dietary segregation of seabirds at the subantarctic Crozet Islands. Marine Ornithology 22:1–192.
Roe H.S.J. and Badcock J. (1984) The diel migrations and distributions within a mesopelagic community in the North East Atlantic. 5. Vertical migrations and feeding of fish. Progress in Oceanography 13:389–424.[CrossRef][Web of Science]
Santos M.B., Clarke M.R., Pierce G.J. (2001) Assessing the importance of cephalopods in the diets of marine mammals and other top predators: problems and solutions. Fisheries Research 52:121–139.[CrossRef][Web of Science]
Stillwell C.E. and Kohler N.E. (1985) Food and feeding ecology of the swordfish in the eastern Atlantic with estimates of daily ration. Marine Ecology Progress Series 22:239–247.[Web of Science]
Takahashi M., Okazaki M., Okamura H., Yokawa K. (2003) Swimming behaviour and migration of a swordfish recorded by archival tag. Marine and Freshwater Research 54:527–534.[CrossRef][Web of Science]
Whitehead P.J.P., Bauchot M.-L., Hureau J.-C., Nielsen J., Tortonese E. (1989) In Whitehead P.J.P., Bauchot M.-L., Hureau J.-C., Nielsen J., Tortonese E. (Eds.). Fish of the North-eastern Atlantic and the Mediterranean, vol. 1(UNESCO, Paris) 510 pp.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||