The onset of piscivory in a freshwater fish species: analysis of behavioural and physiological traits

The onset of piscivory in fish, resulting in a shift from zooplankton or invertebrate to fish prey, was studied on pikeperch (Sander lucioperca) larvae displaying and not displaying piscivorous behaviour at different ages (23, 30, 37, 44 and 52 days post-hatching) using behavioural (attack, capture and swimming activity), morphological (allometry) and digestive enzymatic (trypsin, α-amylase and pepsin) analyses. The shift from zooplanktonic food items (Artemia nauplii) to a piscivorous diet did not occur at the same time for all individuals within the same cohort. Predation tests, conducted under controlled conditions (20°C; ad libitum feeding), showed that some larvae attacked fish prey as early as the age of 3 weeks [11.0 ± 1.3 mm total length (TL)], whereas others did not start until the age of 6 weeks (16.6 ± 1.9 mm TL). Piscivorous individuals were bigger, with larger heads, longer tails, higher acid protease and lower alkaline protease activities, than non-piscivorous conspecifics. In conclusion, high interindividual variability in morphological and digestive system developments linked to the development of predatory abilities could induce cannibalism in fish.     

Diet and trophic niche of Antarctic silverfish Pleuragramma antarcticum in the Ross Sea,Antarctica

The diet of Antarctic silverfish Pleuragramma antarcticum was evaluated by examining stomach contents of specimens collected in the Ross Sea (71°–77° S; 165°–180° E) in January to March 2008. Pleuragramma antarcticum (50–236 mm standard length, L_S) and prey items were analysed for stable‐isotopic composition of carbon and nitrogen. According to index of relative importance (I_RI), which incorporates frequency of occurrence, mass and number of prey items, the most important prey items were copepods (81%I_RI over all specimens), predominantly Metridia gerlachei and Paraeuchaeta sp., with krill and fishes having low I_RI (2·2 and 5·6%I_RI overall). According to mass of prey (M) in stomachs, however, fishes (P. antarcticum and myctophids) and krill dominated overall diet (48 and 22%M, respectively), with copepods being a relatively minor constituent of overall diet by mass (9·9%M). Piscivory by P. antarcticum occurred mainly in the extreme south‐west of the region and near the continental slope. Krill identified to species level in P. antarcticum stomachs were predominantly Euphausia superba (14·1%M) with some Euphausia crystallophorias (4·8%M). Both DistLM modelling (PRIMER‐permanova+) on stomach contents (by I_RI) and stepwise generalized linear modelling on stable isotopes showed that L_S and location were significant predictors of P. antarcticum diet. Postlarval P. antarcticum (50–89 mm L_S) consumed exclusively copepods. Juvenile P. antarcticum (90–151 mm L_S) consumed predominantly krill and copepods by mass (46 and 30%M, respectively). Small adult P. antarcticum (152–178 mm L_S) consumed krill, fishes and copepods (37, 36 and 15%M, respectively). Large adult P. antarcticum (179–236 mm L_S) consumed predominantly fishes and krill (55 and 17%M, respectively), especially in the north (near the Ross Sea slope) and in the SW Ross Sea. Amphipods were occasionally important prey items for P. antarcticum (western Ross Sea, 39%M). General concordance between stomach contents and trophic level of P. antarcticum and prey based on δ¹⁵N was demonstrated. Pleuragramma antarcticum trophic level was estimated as 3·7 (postlarval fish) and 4·1 (fish aged 3+ years).     

Ontogenetic diet shifts of rough scad Trachurus lathami in the North Patagonian shelf (south-west Atlantic Ocean)

Rough scad Trachurus lathami is a key pelagic fish in the Argentinean continental shelf (ACS, south-west Atlantic Ocean), with recent increases in abundance. It is a main prey of fishes and marine mammals, and shares the environment with commercially relevant pelagic species (Engraulis anchoita and Scomber colias), playing an important role linking lower and upper trophic levels in the ecosystem. This study aims to determine the ontogenetic changes in the diet composition, feeding strategy, trophic niche breadth and trophic level of T. lathami in the North Patagonian Shelf (43°–45°30′S). The stomach contents of adult fish (n = 238) were analysed. The results suggest a clear ontogenetic shift in the diet at a size of ~190 mm. Smaller individuals (160–190 mm) were specialized on misidaceans, and showed the highest trophic level, while larger T. lathami (221–230 mm) consumed decapods (Peisos petrunkevitchi) and teleosts (eggs and larvae). Trophic niche breadth was higher at the medium-sized class (191–220 mm), which mainly preyed on copepods (Calanoides carinatus) and chaetognaths (Sagitta spp.), evidencing a more diverse diet and a rather generalist strategy. Updated information on the trophic ecology of T. lathami evidences its extremely plastic feeding behaviour, being able to adapt its trophic niche to the most readily available food items from the mesopelagic community.     

Feeding convergence in South American and African zooplanktivorous cichlids Geophagus brasiliensis and Tilapia rendalli

Synopsis Acará, Geophagus brasiliensis, and red-breasted bream, Tilapia rendalli, are important planktivorous cichlids in southern Brazilian lakes and reservoirs. In laboratory experiments, I quantified behavior and selectivity of different sizes of these two fish feeding on lake zooplankton. Feeding behavior depended on fish size. Fish < 30 mm were visual feeders. Fish 30–50 mm either visually fed or pump-filter fed depending on zooplankton size. Fish > 70 mm were pump-filter feeders. Replicate 1 h feeding trials revealed that, as the relative proportions of prey changed during an experiment, acará (30–42 mm, standard length) and tilapia (29–42 mm) shifted from visual feeding on large evasive copepods to filter feeding on small cladocerans and rotifers. Electivity and feeding rate increased with prey length, but were distinct for similar-sized cladocerans and copepods. Visual/filter-feeding fish had lowest electivities for small and poorly evasive rotifers and cyclopoid nauplii. They fed non-selectively on cyclopoid copepodites, had intermediate electivities for calanoid nauplii and small cladocerans, and had highest electivities for large cladocerans, cyclopoid adults, and calanoid copepodites and adults. Although belonging to different cichlid genera and native to South America and Africa, respectively, acará and red-breasted bream (= congo tilapia) exhibited similar selectivity for zooplankton. Apparently, few stereotyped feeding behaviors have evolved during the acquisition of microphagy in fish. Shift in feeding modes allows these two species to optimally exploit the variable and dynamic patchy distribution of planktonic resources.     

Feeding patterns of post-larval and juvenile notothenioids in the southern Weddell Sea (Antarctica)

Summary The food of 163 juvenile specimens of 13 species of notothenioid fishes collected in the southern Weddell Sea (Antarctica) was analyzed. Investigated fish size range was 3–13 cm SL. Principal food items were calanoid copepods Metridia gerlachei, Calanoides acutus, and Calanus propinquus; all developmental stages of Euphausia crystallorophias, and post-larval nototheniid fish Pleuragramma antarcticum. Diet of juvenile channichthyids is limited to few species of euphausiids and fish in the size > 10 mm, but does not include significant numbers of copepods. Pelagic stages of nototheniids feed on copepods and/or larval euphausiids smaller than 10 mm. At similar size, nototheniids and bathydraconids take smaller prey items than channichthyids.     

Feeding strategies of Atherina boyeri (Risso 1810) in a freshwater ecosystem

This study investigates the adaptation of certain feeding strategies by a population of Atherina boyeri inhabiting a freshwater ecosystem (Lake Trichonis, Greece) and links feeding habits with ecosystem interactions, using stomach content analysis and food availability. The use of Ivlev's electivity index (E) revealed the selection of different prey categories in respect to size, species and even gender. A. boyeri showed positive selection for larger zooplankters such as adult copepods and the egg‐bearing females of Eudiaptomus drieschi (0.53 < E < 0.92 and 0.24 < E < 0.99, respectively) or the cladocerans Diaphanosoma orghidani and Daphnia cucullata (0.25 < E < 0.53 and 0.23 < E < 0.37, respectively) in spring and/or summer. In contrast, the fish totally avoided rotifers and copepod nauplii, while showing negative selection for smaller prey items like copepodites (?1.00 < E < ?0.14). A shift to larger prey was recorded as age increased. Larger specimens showed cannibalistic behaviour by consuming larvae of A. boyeri and the endemic, threatened, fish Economidichthys trichonis. The results provide certain indications that, through selective predation, A. boyeri is able to affect zooplankton abundance and community composition, the vertical distribution and migration of crustaceans and the abundance of other fish species, However, more research on the dynamics and feeding ecology of A. boyeri and other fish species is necessary.     

Prey Detection and Prey Capture in Copepod Nauplii

Copepod nauplii are either ambush feeders that feed on motile prey or they produce a feeding current that entrains prey cells. It is unclear how ambush and feeding-current feeding nauplii perceive and capture prey. Attack jumps in ambush feeding nauplii should not be feasible at low Reynolds numbers due to the thick viscous boundary layer surrounding the attacking nauplius. We use high-speed video to describe the detection and capture of phytoplankton prey by the nauplii of two ambush feeding species (Acartia tonsa and Oithona davisae) and by the nauplii of one feeding-current feeding species (Temora longicornis). We demonstrate that the ambush feeders both detect motile prey remotely. Prey detection elicits an attack jump, but the jump is not directly towards the prey, such as has been described for adult copepods. Rather, the nauplius jumps past the prey and sets up an intermittent feeding current that pulls in the prey from behind towards the mouth. The feeding-current feeding nauplius detects prey arriving in the feeding current but only when the prey is intercepted by the setae on the feeding appendages. This elicits an altered motion pattern of the feeding appendages that draws in the prey.     

Feeding ecology of the Bay Goby (Pisces: Gobiidae). Effects of behavioral,ontogenetic, and temporal variation on diet

The ontogenetic, diel, seasonal, and yearly variations in gut fullness, diet, and prey diversity for a California estuarine gobiid (Lepidogobius lepidus Girard) were examined. Also the feeding behavior of this species was described.Small (<50 mm, SL) and large (? 50 mm, SL) gobies consumed similar prey types in different proportions. Major prey items were polychaetes, harpacticoid copepods, gammarid amphipods, molluscs, and other crustaceans. Diets of large and small gobies were not significantly correlated, and larger fish had a more diverse diet. Small fish fed at all times while larger gobies fed primarily at night. Changes in diet may be related to differential prey preferences, feeding chronologies, and increases in fish size.Both large and small gobies displayed seasonal differences in diet and prey diversity. Year-to-year changes in diet also were noted for both size classes. The bay goby uses different feeding behaviors to capture sedentary and motile prey and appears to forage opportunistically. This behavior is probably advantageous in an environment which fluctuates drastically.     

Feeding ecology of juvenile turbot Scophthalmus maximus and flounder Pleuronectes flesus at Gotland, Central Baltic Sea

Food and feeding of juvenile turbot Scophthalmus maximus and flounder Pleuronectes flesus were studied in five nursery areas at Gotland, Central Baltic Sea, ICES SD 27 and SD 28. Ontogeny involved partitioning of available food resources. The food choice of turbot <30 mm standard length (L_S) included both planktonic‐hyperbenthic prey (calanoid copepods and mysids) and epibenthic–endobenthic prey (chironomids and amphipods), whereas turbot ≥30 mm L_S fed mainly on hyperbenthic species (mysids and fishes). Conversely, for flounder, epibenthic–endobenthic prey were the most abundant prey items throughout development (harpactocoid copepods, oligochaetes and chironomids for fish <40 mm L_S and oligochaetes, chironomids and amphipods for flounder ≥40 mm L_S). Thus, the highest degree of dietary overlap occurred between turbot <30 mm and flounder ≥40 mm. Food composition for both turbot and flounder varied, however, according to exposure and predominant wind direction in the nursery area. For example, expressed as the ratio between the biomass of mysids and fishes consumed, the relative importance of mysids v. fishes as food source for turbot, varied from <1 in the most sheltered area to 16 and 27 in the more open areas. Considerable differences in feeding incidence were recorded; mean ±s .d . 58 ± 20% for turbot <30 mm L_S and 83 ± 8% for turbot ≥30 mm L_S, as opposed to ≥85–90% for flounder irrespective of size. The lower feeding success of turbot <30 mm L_S was related to mysid abundance, shown to vary spatially and temporally, and to density of flounder, indicating that food availability, and potentially interspecific competition, influence feeding of early juvenile turbot with implications for survival following settlement. Regarding variability in abundance, hyperbenthic prey, as mysids, are considered more variable than epi‐ and endobenthic organisms. Hence, in addition to the ‘nursery size hypothesis’, i.e. the positive relationship between abundance of recruits and extension of nursery areas, variability in food availability may explain the average lower recruitment of turbot as compared to other flatfishes, e.g. flounder.     

Food of an endangered cyprinodont (Aphanius iberus): ontogenetic diet shift and prey electivity

Synopsis We studied the ontogenetic diet shift and prey electivity of an endangered cyprinodontid fish endemic to the Iberian Peninsula, the Spanish toothcarp (Aphanius iberus). The toothcarp’s diet was omnivorous, dominated by harpacticoid copepods (Mesochra lilljeborgi and Tisbe longicornis), copepod nauplii and detritus. Diet composition varied greatly among habitats, depending on prey availability. In a rarely inundated habitat (glasswort), there was more consumption of the isopod Protracheoniscus occidentalis and the harpacticoid copepod Mesochra lilljeborgi, while in algal mats another harpacticoid (Tisbe longicornis), chironomid dipterans and invertebrate eggs were more important in diet. Although a benthic feeding habitat has previously been suggested, in our study the diet was based rather on water column organisms for both glasswort and algal mat habitats. There was also an ontogenetic diet shift, with an increase of mean prey length with fish length, clearly linked to a microhabitat change. Smaller fish showed positive electivity and greater reliance on planktonic prey (e.g. copepod nauplii, the harpacticoid copepods Mesochra lilljeborgi and Tisbe longicornis, the rotifer Brachionus plicatilis, and ostracods), while larger fish elected and preyed on more benthic organisms (e.g. Canuella perplexa, Mesochra rapiens, and ephydrid dipterans).     

Temporal variations in the diet of pearly razorfish Xyrichtys novacula (Osteichthyes: Labridae)

The diet of pearly razorfish Xyrichtys novacula, caught monthly along the shores of the Island of Lampedusa, appeared to be mainly composed of crustaceans, followed by colonial ascidians, molluscs and polychaetes. Among prey, sand dwellers and phanerogam‐associated species were recorded. In winter months, the diet was characterized by a small number of prey items, dominated by colonial ascidians, while in spring and summer a wider prey array was recorded. Dietary indices show that X. novacula do not strictly exploit benthic prey but also pelagic organisms, such as copepods. This feeding behaviour reached its peak in March and October, when the abundance of primary consumers was at its highest after phytoplankton blooms. Furthermore, X. novacula caught prey organisms according to their availability and seasonal patterns during their life cycles, irrespective of fish size.     

Feeding by larvae of Hypoatherina tropicalis (Pisces: Atherinidae) and its relation to prey availability in One Tree Lagoon,Great Barrier Reef,Australia

Synopsis Feeding of and food availability for larvae of Hypoatherina tropicalis were investigated in One Tree Lagoon, Great Barrier Reef, Australia, during November 1981 and January 1982. These surface-dwelling larvae and their microzooplankton prey were sampled as near to simultaneously as possible on 12 occasions during the daytime. Larvae of all sizes (5–17 mm SL) fed successfully over the observed range of mean prey densities (12–235 per liter), and the overall feeding incidence was 98.9%. Larger larvae consumed greater numbers and more categories of prey than did smaller larvae. Larvae selected copepods of all sizes, and nauplii, gastropods, bivalves, and foraminiferans that were greater than 75 ¢s mm in width. Tintinnids (mostly 37–74 µm in width) were generally avoided by larvae, but were occasionally important in the diets when they constituted more than 60% of the total available prey, regardless of the density of the selected prey categories. Larvae less than 14 mm SL ingested meroplankton (gastropods, bivalves, foraminiferans, and polychaetes) in direct relation to the densities available, and without regard to the densities of copepods available. However, the largest larvae (14–17 mm SL) ingested meroplankton in inverse relation to the density of copepods available, indicating that larvae consumed more meroplankton when the concentration of copepods was low. Such flexibility and opportunism in feeding behavior may increase the larvae's chances of obtaining adequate nutrition during periods of suboptimal feeding conditions.     

Intertidal recruitment and feeding in relation to optimal utilization of nursery areas by juvenile English sole (Parophrys vetulus: Pleuronectidae)

Synopsis Juvenile Parophrys vetulus 19–102 mm were collected on an intertidal flat in Humboldt Bay during 1976–1977. Recently metamorphosed fish that fed almost exclusively on harpacticoid copepods and other epibenthic crustaceans were found in large numbers during spring and early summer. A feeding transition occurred among fish 50–65 mm in length, and infaunal polychaetes were the dominant prey of fish greater than 65 mm. Increases in the feeding niche width and average size of prey items, as well as a decrease in the number of prey items per stomach, accompanied growth.The disappearance of English sole from intertidal areas in early fall at an average size of 82 mm and their subsequent residence in subtidal channels until a size of 140 mm suggest a size-depth segregation within the nursery ground. The advantages of such a distribution are discussed in terms of optimal foraging and reduction of intra- and interspecific competition.     

Bioenergetic model of planktivorous fish feeding, growth and metabolism: theoretical optimum swimming speed of fish larvae

The feeding activity of an individual fish larva is described by an equation which includes parameters for the area successfully searched, probability of food capture multiplied by the cross-sectional perceptive visual field, larval swimming speed and the time required to consume a unit of food energy. The proportion of ingested food energy used for metabolism increases exponentially with increasing swimming speed. The model predicts that food consumption rate increases asymptotically whereas metabolic rate increases exponentially. This results in a predicted growth rate curve that reaches a maximum at a certain swimming speed and decreases at both higher and lower speeds. The model can be used to predict the influence of type of prey, prey density, water temperature etc. on larval growth. An expression describing how many hours per day fish larvae must forage in order to grow at a certain daily body weight gain allows the limits of environmental conditions for positive, zero and negative growth rate to be set. Results of simulations demonstrated that the optimum swimming speed for maximum growth of coregonid larvae increased with an increase in food density, decrease in water temperature or decrease of prey vulnerability. At optimum ‘theoretical’ swimming speed an increase in water temperature from 5 to 17° C required the food density to be increased from 20 to 80 copepods l^?1 in order to maintain a daily growth increment of 2%. The minimum Artemia density required for maintenance metabolism increased from 10 to 30 items 1¹ over the same temperature increase from 5 to 17° C, and food densities required for 8% growth rates were 26 and 56 Artemia nauplii l^?1 at 5 and 17° C, respectively. Contrary to previous findings, results of the present study suggest that metabolic rates of actively feeding fish larvae may be from 5 to 50 times the standard metabolic rate: earlier studies suggested that a factor of 2–3 may be generally applicable.     

Rearing tench (Tinca tinca L.) larvae on live feed (Artemia) and on two transition schedules from live to dry diets

Two 60‐day experiments were carried out on tench (Tinca tinca L.) from day 5 post‐hatch. Density was 20 larvae L^?1 and temperature 24 ± 0.5°C. In experiment 1, Artemia nauplii were the sole food, testing nauplii amounts and feeding frequency. High survival rates (between 79.5% and 95.5%) were obtained. Growth was faster as nauplii amounts were greater; the highest growth rate (11.00), weight (265.5 mg) and Fulton’s coefficient (1.40) were obtained when fish were fed in excess once a day, without significant differences from the growth obtained by feeding in excess of eight times a day. In experiment 2, a dry diet for marine fish was tested as a replacement for Artemia nauplii, following two transition protocols, one faster than the other; high survival rates (between 77.7% and 87.1%) were again obtained. The slower transition allowed a growth rate of 10.14, length of 23.1 mm, weight of 158.3 mg and a Fulton’s coefficient of 1.28, without significant differences from the faster transition. At all stages, growth values were significantly higher from feeding nauplii in excess as the sole food, but the required nauplii quantity was six times higher than the amount supplied to the animals fed the dry diet.     

Appendicularians: an important food supply for the Argentine anchovy Engraulis anchoita in coastal waters

The stomach contents of 532 adult specimens of the Argentine anchovy, Engraulis anchoita, caught in coastal waters of the Argentine Sea from 1994 to 1996, were analyzed. Larger amounts of food were found in the stomachs of anchovies collected in the northern surveys (35–40°S), with the highest values attained in 1994. Main diet items were copepods, appendicularians, cladocerans, fish eggs, and pteropods. Anchovies caught in the southern surveys (40–45°S) ingested less food, whereby the main items were copepods and appendicularians. Plankton samples collected simultaneously with the fishing trawls were also analyzed. Except for Fritillaria borealis, which occurred only in plankton samples, the same appendicularian species (Oikopleura dioica and O. fusiformis) were found in both stomach contents and in plankton samples. The Ivlev selectivity index calculated for O. dioica and O. fusiformis and for the different maturity stages of both species were in all cases ～0, supporting the hypothesis of a non‐selective feeding.     

Prey selection by larvae of Prochilodus lineatus (Pisces: Curimatidae): indigenous zooplankton versus veligers of the introduced bivalve Limnoperna fortunei (Bivalvia: Mitilidae)

We studied experimentally the feeding selectivity of larvae of Prochilodus lineatus (Pisces), with particular emphasis on the role of veligers of the exotic bivalve Limnoperna fortunei. Three concentrations of veligers were offered to three developmental stages of P. lineatus. Veliger concentrations were: (1) higher than in the field (“enriched”, 0.09 ind. ml⁻¹), (2) unmodified from field conditions (“normal”, 0.06 ind. ml⁻¹), and (3) lower than in the field (“low”, 0.02 ind. ml⁻¹). Fish developmental stages were protolarvae (approx. 10 days old), mesolarvae (17 days), and metalarvae (25 days). Proportions (in terms of numbers and biomass) and selectivity values were calculated for each prey item evaluated: veligers, small cladocerans + nauplii, medium-sized cladocerans, copepodits, and large cladocerans + copepods. Protolarvae and mesolarvae consumed veligers almost exclusively (88–90%, both in numbers and in biomass) when offered prey enriched in veligers, whereas for metalarvae veligers represented only 16.0% of the food consumed. At lower veliger concentrations, only protolarvae preferred Limnoperna veligers, whereas older fishes switched gradually to crustacean plankton. We conclude that veligers are preferred by the early fish developmental stages, and we speculate that this may be because their slower swimming makes them easier to capture than planktonic crustaceans. However, as fish larvae grow larger, veligers become too small a prey for their energetic needs, and they switch to larger items like cladocerans and copepods. We anticipate that this new and abundant food resource has an important impact on the survival and growth of P. lineatus.     

Dietary overlap among early juvenile stages in an Antarctic notothenioid fish assemblage at Potter Cove,South Shetland Islands

To date, studies of food overlap in Antarctic fish have been performed on a mixture of late juvenile and adult stages, leaving the young immature specimens (TL ≤ 10 cm) practically unexplored. We studied diet overlap and potential competition among early juvenile individuals in a coastal notothenioid community at Potter Cove, by analysing the stomach contents of 225 fish of 5 species collected in the summer of 2009–2010. We used frequency of occurrence (F %) and the coefficient “Q” for diet evaluation and the method of Tyler and the similarity index “S” for food overlap. Amphipods of the suborder Gammaridea were the main (Q > 2.900) and most frequent (% F) prey for all species, although Notothenia coriiceps also consumed gastropods of the family Littorinidae, mostly Laevilitorina antarctica. Secondary prey were algae for Notothenia rossii and N. coriiceps, calanoid (pelagic) and harpacticoid (benthic) copepods for Trematomus newnesi and the latter copepods and isopods of the family Munnidae for Lepidonotothen nudifrons. The reoccurrence of prey among fish species was 39.6 % and food overlap between 90 % of species pairs was under 58 %. Because similarly low values of diet overlap were reported for intermediate/advanced juveniles and adults of the same species at the same site, we conclude that there is no difference in the degree of interspecific food overlap and therefore potential competition between the immature and mature fraction of the fish community. Food competition is avoided by resource partitioning along a depth gradient or by different prey species.     

Effect of zooplankton type and abundance on prey consumption by the fairy shrimp, Streptocephalus proboscideus (Anostraca: Crustacea)

Laboratory cultured Streptocephalus proboscideus (three sizes (mm), viz. 8.44 ± 0.95 (virgin), 14.18 ± 1.49 (adult I) and 19.24 ± 1.52 (adult II)) were offered (separately for males and females) field collected zooplankton (12 prey types) at three levels of abundance (1.0 ml⁻¹, 2.0 ml⁻¹ and 4.1 ind. ml⁻¹ in 30-minute feeding experiments. Gut contents, analyzed for abundance and diversity of prey type, showed that predator size, sex and their interaction had strong effects on prey consumption. Regardless of their size, and of prey density, S. proboscideus females consumed 25–90% more prey than males. Their filtration rates (adult II) were higher (125 ml ind.⁻¹ h.⁻¹) than those of males (30 ml ind.⁻¹ h.⁻¹) too. Rotifers had the highest numerical percentage in the gut, regardless of predator size or sex. Cladocerans were only consumed by adults I and II. Adult II females consumed 28.5–43.3 μg zooplankton dry weight ind.⁻¹ h.⁻¹. Size distribution of B. longirostris in the field and in the gut were closely similar. This study confirms S. proboscideus as a non-selective filter feeder. Since it did not eat jumping rotifers, copepod nauplii and copepodites, it may contribute to structuring its prey communities, because good escapers will be enriched in the medium, while poor escapers will be depleted.