The feeding, growth and behaviour of juvenile cod, Gadus morhua L., in cold environments

Most studies on feeding-related activities of fish in cold water (0° C) have not used eurythermal species or examined a broad suite of activities. In this study we report on the feeding, growth, behaviour and conversion-efficiency of 0 + Atlantic cod in response to water temperature and food availability. In one experiment, activity, opercular beat and growth decreased with decreasing temperature over the range 8·3–0·6° C. However, the conversion efficiency increased with decreasing temperature, a result partly explained by the relative change in activity and opercular rate. In another experiment, temperature had a linear effect on growth in cod fed different food rations. Both food availability and temperature were found to influence growth in group 0 + cod. The specific growth rate of cod in this experiment exceeded 2% day^-1. Growth data collected from cod sampled in the field during the cold-water period showed a progressive increase of mean length over the season. Overall, these results demonstrate that 0 + cod have adapted to life in cold-water environments, in contrast to adult cod which mostly inhabit warmer environments.     

Ontogenetic variations of thermal optimum for growth, and its implication on thermolabile sex determination in blue tilapia

Knowledge of how the optimum temperature for growth ( T °_opt) varies during ontogeny, and how close it is to the temperatures that induce phenotypic masculinization is fundamental to the understanding of the evolution of thermolabile sex determinism (TSD) in fishes. In blue tilapia Oreochromis aureus , T °_opt is 32·6° C at the start of exogenous feeding (10mg fish) and it decreases by c . 1° C each time that the fish body mass increases by an order of magnitude. Temperatures <35° C are not sufficient to induce complete phenotypic masculinization. Based on a multiple-regression model ( r ²=0·938) plotting growth against body mass and water temperature, genotypically female tilapia living at high temperatures during the thermosensitive period (21–28 days) and being reversed into phenotypic males would incur an initial growth disadvantage over fish living at T °_opt, but not over those living at slightly colder temperatures (27–29° C). This initial disadvantage would be later compensated for by faster growth because of between-sex growth dimorphism to the detriment of phenotypic females. These arguments suggest that there is no definite pressure against the selection of TSD in blue tilapia and probably other Oreochromis spp.     

Microhabitat selection in the early juvenile mudskipper Boleophthalmus pectinirostris (L.)

Using laboratory choice experiments, behavioural preferences of the early juvenile mudskipper Boleophthalmus pectinirostris for temperature, salinity and sediment were observed. The temperature preference experiments were conducted in an annular chamber with a thermal gradient of 27 to 34° C, and the fish selected a mean ± s . d . temperature of 31·2 ± 0·5° C. The salinity preference experiments were conducted in an aquarium with decreasing salinity from 20 to 15, 10, 5 and 0·5, and significantly more ( P < 0·05) fish were found in water with a salinity of 5. The sediment preference experiments were tested in circular tanks which contained four types of sediment (medium sand, fine sand, muddy sand and sandy mud), and the fish showed a clear preference for sandy mud. These results indicated that early juvenile B. pectinirostris showed behavioural preference for microhabitats. Temperature and salinity probably set large-scale boundaries on distribution, but sediment should be a critical factor for determining the distribution of the mudskipper.     

Plasma cortisol and 17β-oestradiol levels in roach exposed to acute and chronic stress

Plasma cortisol levels were measured as an indicator of physiological stress in roach subjected to brief handling, or to a 14-day period of confinement, and in undisturbed control fish, during winter (water temperature 5° C) and summer (16° C), at which time plasma 17 β-oestradiol levels were also determined. Cortisol levels in undisturbed roach were low (mean 8·1 ng ml⁻¹ at 5° C; 1·4 ng ml⁻¹ at 16° C) and both handling and handling+confinement elevated blood cortisol levels significantly to 400 and 140 ng ml⁻¹, respectively (at 5° C) and 700 and 600 ng ml⁻¹, respectively (at 16) C). Blood cortisol levels had almost returned to baseline within 4 h following handling alone but in fish subjected to handling and prolonged confinement cortisol levels remained elevated for up to 168 h. Differences in baseline and poststress levels of cortisol, and in the rate of recovery from acute stress, were observed at the two different temperatures and the possible factors underlying these differences are discussed. Circulating levels of 17 β-oestradiol were reduced significantly within 24 h of exposure to either acute handling or chronic confinement indicating that the reproductive endocrine system in roach is sensitive to disruption by stressors.     

Temperature-induced changes of survival, development and yolk partitioning in Chondrostoma nasus

Fertilized Chondrostoma nasus eggs were incubated at 10, 13, 16 and 19° C until full resorption of the yolk sac. High survival was observed at 10–16° C (89–92% at the onset of external feeding), whereas at 19) C survival was depressed (76%). The time at which 5, 50 and 95% of individuals had hatched, filled the swim bladder, ingested the first food and fully resorbed the yolk sac was determined. An increase in temperature accelerated development and made it more synchronous. Within the period from fertilization to hatching embryonic development was theoretically arrested (t_{0 dev}) at 8·8° C, and growth was arrested (t_0gr) at 8·86° C. For the whole endogenous feeding period (from fertilization to full yolk resorption) the amount of matter transformed into tissue was temperature independent between 10° and 19° C. Respiration increased exponentially with age; the respiration increase was faster at higher temperatures, but, in general, metabolic expenditures of C. nasus were low. As a consequence, the efficiency of utilizing yolk energy for growth was high as compared with other fish species (57% during the whole endogenous feeding period); it was temperature independent. However, time was used less efficiently at low temperatures, increasing a risk of predation. Within the endogenous feeding period a shift from lower to higher temperatures for optimal yolk utilization efficiency was observed. The temperatures optimal for survival and energetic performance seem to be 13–16° C for egg incubation and 15–18° C for rearing of yolk-feeding larvae. Chondrostoma nasus is a potential candidate for aquaculture for restocking purposes.     

Effects of temperature, body size and feeding on rates of metabolism in young‐of‐the‐year haddock

The mean rate of oxygen consumption (routine respiration rate, R _R, mg O₂ fish⁻¹ h⁻¹), measured for individual or small groups of haddock Melanogrammus aeglefinus (3–12 cm standard length, L _S) maintained for 5 days within flow‐through respiratory chambers at four different temperatures, increased with increasing dry mass ( M _D). The relationship between R _R and M _D was allometric ( R _R = α  M ^b ) with b values of 0·631, 0·606, 0·655 and 0·650 at 5·0, 8·0, 12·0 and 15·0° C, respectively. The effect of temperature ( T ) and M _D on mean R _R was described by     indicating a Q ₁₀ of 2·27 between 5 and 15° C. Juvenile haddock routine metabolic scope, calculated as the ratio of the mean of highest and lowest deciles of R _R measured in each chamber, significantly decreased with temperature such that the routine scope at 15° C was half that at 5° C. The cost of feeding ( R _SDA) was c . 3% of consumed food energy, a value half that found for larger gadoid juveniles and adults.     

The influence of previous feeding regime on the compensatory growth response of maturing and immature Arctic charr, Salvelinus alpinus

Alternating periods of food deprivation with those of unlimited provision of food depressed the growth of Arctic charr, Salvelinus alpinus , below that of controls. Fish that were deprived of food and then fed on alternate weeks (1:1) were larger than those that were exposed to periods of 1 5- or 3-week deprivation and feeding (1·5:1·5 or 3:3). On receiving excess food supplies following 24 weeks on the restricted feeding regimes the previously-restricted fish grew more rapidly than the controls. The greatest compensatory growth was displayed after the 3:3 regime, followed by the 1·5: 1·5 and then the 1:1 feeding regime. At the termination of the experiment there were no significant differences in body weight between fish fed according to the different regimes during the period that food restriction was imposed. Growth patterns of the immature males and females were similar, but mature males were significantly lighter than the immature fish by the end of the experiment. Both immature and maturing fish displayed a compensatory growth response on return to adequate feeding. Beginning food restriction in May did not influence the proportions of male fish ( c . 60%) which were mature in the autumn.     

Upper thermal limits for feeding and growth of 0+ Arctic charr

When Arctic charr Salvelinus alpinus from two diVerent stocks were fed live Neomysis integer , the upper thermal limits for feeding and growth were established in the range 21·5–21·8° C. These critical temperatures might have been underestimates, because fish tend to show increased sensitivity to handling at high experimental temperatures. In the second experiment, the proportion of feeding undisturbed charr from four stocks decreased initially as temperature was raised in steps from 18 to 22° C. At the lower temperatures, 18 and 20) C, almost all fish resumed feeding, but the recovery time was longer and more fish ceased to feed at 20) C than at 18° C. When the temperature was increased to 21° C, 50% of the fish ceased feeding permanently, and all fish ceased feeding within 2 days at 22° C. It is concluded that 0+ charr cease to feed and grow at c .21·5) C and that the critical temperatures for feeding and growth coincide.     

Toxicity of rotenone to some species of coarse fish and invertebrates

The toxicity of rotenone was determined for seven species of fish and three species of invertebrates. The results show that there is a marked variation in sensitivity of different species to rotenone. The survival times of roach Rutilus rutilus (L.) were reduced by a rise in temperature and also by a reduction in water hardness. Temperature affected the rate at which rotenone is degraded; 3 days after being prepared a solution containing 2 mg/1 of 5 % rotenone was non-toxic to roach over a 7-day period at 20°C, but it was still toxic for at least 11 days at 11.5°C. The data are discussed in relation to the use of rotenone in fisheries management.     

Behavioural studies on the lesser sandeel Ammodytes marinus (Raitt) III. The effect of temperature on activity and the environmental control of the annual cycle of activity

The behaviour of the lesser sandeel, Ammodytes marinus (Raitt), has been investigated at 5, 10 and 15° C, using a photographic method of recording activity. The activity patterns at 10 and 15° C were very similar, there being a high level of swimming activity during the light period, which fell to a low level at 5° C. It was also lower at 10° C at the end of the experiment than at the beginning and it is suggested that this might have been due to an increase in the fat contents of the fish. The feeding rate of the fish was measured and showed a Q ₁₀ of 2.08 for the temperature range 5–15° C. The annual cycle of activity of A. marinus is discussed in relation to seasonal changes in food availability, light and temperature, and in the fat content of the fish. It is concluded that after spawning in the December–January period the fish remain buried in the sand until April, because of the limiting effect on swimming and feeding activity of the environmental factors in the intervening period. The proportion of fish available for capture at the start of the fishery in April is related mainly to temperature, but food (as measured by numbers of copepods) light intensity and photoperiod are by then increasing rapidly. After July the fishery ceases and it is thought that this is because the fish have entered an overwintering stage, during which they remain buried in the sand. This phase is also thought to be associated with the maturation of the gonads in readiness for the winter spawning. The factors causing the fish to enter this stage are as yet undetermined but may be related to the attainment of a certain level of fat content.     

The adaptation of digestive enzymes to temperature, season and diet in roach, Rutilus rutilus and rudd Scardinius erythrophthalmus; Proteases

The seasonal changes of proteolytic activity in the gut content of roach R. rutilus and rudd S. erythrophthalmus in four Tyrolean lakes and the adaptation of the proteases to constant temperatures and different natural diets were studied. In rudd proteolytic activity remains nearly constant throughout the year. In roach proteolytic activity increases and then decreases during the first three months after the thaw in spring. This period appears to be endogenously controlled and is followed by a second stage in which proteolytic activity is influenced by environmental factors. Under natural conditions both species have a higher proteolytic activity when feeding on animals than on detritus. Under laboratory conditions there is a slight maximum at an environmental temperature of 16° C when the fish had been fed on meal worms. With plant food a significant dependence of proteolytic activity on environmental temperature was only found in roach.     

Influence of rearing temperature during the larval and nursery periods on growth and sex differentiation in two Mediterranean strains of Dicentrarchus labrax

European sea bass Dicentrarchus labrax of the north‐western (NW) and south‐eastern (SE) Mediterranean Sea strains were exposed to different temperatures (13, 17 or 21° C) during the larval rearing (11–51 days post hatching, dph) or nursery periods (55–95 dph), in order to examine the effects of temperature on sex differentiation and subsequent growth during the first year of life. Higher growth was observed during exposure to higher temperatures, but fish of the NW strain exposed to 13 or 17° C during larval rearing exhibited compensatory growth once exposure to the lower temperatures finished, and as a result their final size at 300 dph was similar or greater to the group exposed to 21° C. Fish exposed to 17° C during the nursery period also had similar size to fish exposed to 21° C after 300 days of rearing, but the fish exposed to 13° C remained significantly smaller (ANOVA, n  = 55–100, P  < 0·05). There were significant differences in the sex ratio among the fish exposed to different temperatures during the two periods of rearing, with high temperature (21° C) resulting in a significantly higher percentage of males in the population, both in the NW (ANOVA, n  = 2, P  < 0·04) and SE populations (ANOVA, n  = 2, P  < 0·01). The masculinization effect of high temperature was significantly stronger during the larval rearing stage, both in the NW (ANOVA, n  = 2, P  < 0·005) and SE populations (ANOVA, n  = 2, P  < 0·01). None of the temperature manipulations could produce 100% females, suggesting that there is a part of the genetic component in sex differentiation which is not labile to environmental influence.     

Rate of energy depletion and overwintering mortality of juvenile walleye pollock in cold water

The winter energy deficit and mortality of juvenile walleye pollock at extremely cold temperature were examined by field observations and laboratory experiments. In the Doto area, along the northern coast of Japan, juvenile walleye pollock resided on the continental shelf despite extremely cold temperatures (mean 0·4° C) during the latter half of winter (March to April). Measurements of the rate of energy depletion (equivalent to the routine metabolic rate) revealed that juvenile walleye pollock consumed 37% less energy at 0·5° C than at 2·0° C, suggesting an energetic benefit of residence in cold water (<1·0° C) over the shelf during winter. Prior to the starvation experiments, temperatures and ration level in the holding tanks were adjusted to create two different body condition groups of fish. Under the thermal condition of the latter half of winter (0·5° C), fish with a mean condition factor of 0·6 and 0·5 suffered 19·1 and 74·5% mortality, respectively, at the end of the experiments (after 56 days). The residual analysis of total body energy demonstrated that the cause of mortality was mainly associated with the depletion of energy reserves. When a logistic regression model for mortality derived from the experiments was applied to wild fish collected in March, the estimated overwintering mortality in 2004 and 2005 was 25·4 and <2·3%, respectively, assuming no feeding during the winter. Considering that juvenile walleye pollock feed during winter as shown in previous studies, intense overwintering mortality induced by energy depletion is improbable during the latter half of winter in the Doto area.     

Reproductive sensitivity to elevated water temperatures in female Atlantic salmon is heightened at certain stages of vitellogenesis

In order to compare the effects on reproductive performance of short-term or prolonged exposure to elevated temperatures during vitellogenesis, female Atlantic salmon Salmo salar were held at a water temperature of 22° C for periods of 4 or 6 weeks during the austral summer and autumn. Plasma levels of 17β-oestradiol (E₂), testosterone (T) and vitellogenin (Vtg) were monitored and reproductive success was compared to that in groups of fish maintained at 14 or 22° C for 12 weeks from mid-January. Significant endocrine effects were observed within as few as 3 days of the commencement of exposure to 22° C, when plasma levels of E₂ ( c. 0·5 ng ml⁻¹) and Vtg ( c. 1·4 mg ml⁻¹) were approximately half those observed in fish maintained at 14° C ( c. 1·0 ng ml⁻¹ and 2·7 mg ml⁻¹ respectively). The fertility and survival to the eyed stage of ova from fish held at 14° C exceeded 85 and 70% respectively, whereas ova from fish held at 22° C for 6 or 12 weeks exhibited significantly reduced fertility (<70 and <45% respectively) and survival ( c. 40 and 13% respectively). In spite of significant endocrine effects at all stages, a 4 week exposure to 22° C only generated significant reductions in egg fertility (<65%) and survival ( c. 30%) when it occurred between mid-February and mid-March. Together, these data confirm that high temperature spikes can affect reproductive success as strongly as more prolonged exposures, and indicate that there is a critical period of reproductive sensitivity to elevated temperature in late February and early March in this stock of Atlantic salmon.     

Energetic costs at different levels of feeding in juvenile cod, Gadus morhua L.

The energetic costs associated with feeding by juvenile cod were determined by means of an open-circuit respirometer. Fish acclimated to several temperatures (7, 10, 15 and 18°C) were kept at natural lighting levels, and fed inside their individual respirometers. They consumed a diet compounded from natural foods, at five different ration levels, their oxygen consumption being monitored continually over an 11–16 day period.
After each meal the rate of oxygen consumption increased to above the pre-feeding level, reaching a peak 8–10 h later. With each successive meal the oxygen consumption showed a cumulative increase, reaching a maximum usually after the last meal.
The elevation in metabolic rate associated with feeding was dependent upon ration size, increasing linearly as the food intake increased. The effect was also dependent upon temperature; for fish fed to satiation the total energy cost was equivalent to 11.9, 10.9, 16.4 and 17.1% of the ingested energy at 7, 10, 15 and 18°C respectively. For resting satiated fish the rate of oxygen consumption was close to the maximum rate for active fish.     

Ecological consequences of food partitioning for the fish population structure in a eutrophic lake

In the highly eutrophic lake, Frederiksborg Slotssø, the diet composition of the bream (Abramis brama L.) and roach (Rutilus rutilus L.) populations was examined during three periods with different food availability. The length range of bream and roach was 9–34 cm (TL) and 5–18 cm (TL), respectively. The relative food composition was examined for 2 cm and 1 cm length intervals of bream and roach, respectively. During all three periods, bream shifted from benthic cladocerans (Alona sp.) to zooplankton and chironomids within a transitional length of 15.0–20.0 cm. These foodshifts were coupled with a change in feeding behaviour from particulate to filter feeding. The biomass of chironomids was too low to sustain the consumption of larger bream (>20.0 cm) which initiated feeding in the pelagic zone even in periods when the mean length and biomass of the preferred zooplankton, Daphnia cucullata, were low. In contrast to bream, roach fed mainly on zooplankton. With increasing size, roach progressively shifted to larger zooplankton species due to the increasing mesh size of their branchial system. The importance of benthic animals in the diet of roach was minor due to low feeding efficiency on prey buried in the sediment. Detritus appeared in the diet of bream and roach in periods of low availability of animal food items. Feeding on detritus may provide an energetic advantage to bream and roach and increase the carrying capacity for these species in lakes, where detritus is highly abundant. Especially for the larger fish due to the decrease in their relative metabolic demands. However, the ability of bream to filter feed and with increasing size to retain food items smaller than those retained by roach may be the main mechanism for the dominance of bream over roach in highly eutrophic lakes.     

Temperature and state‐dependence of feeding and gastric evacuation in juvenile Pacific halibut

Relationships between nutritional state, behavioural response to prey and gastric evacuation rates were examined in juvenile Pacific halibut Hippoglossus stenolepis feeding on squid. Pacific halibut reared at 2, 6 and 10° C were fasted for 1 or 7 days to generate variation in energetic state. The 7 day fast resulted in measurable declines in condition indices at 10 and 6° C but not at 2° C. At 10° C, all Pacific halibut consumed the first meal offered, but fish previously fasted for 7 days took significantly longer to locate and consume the meal than fish fasted for only 1 day. At 2° C, Pacific halibut fasted for 7 days did not generally consume the first meal offered, but resumed feeding 2·1 days sooner, on average, than fish fasted for only 1 day. The gastric evacuation rate of the squid meal was best described by a power model with near‐exponential curvature ( a  = 1·011). The evacuation rate was strongly temperature‐dependent ( Q ₁₀ = 3·65) but displayed the same degree of variability at each temperature. The evacuation rate in Pacific halibut was not affected by feeding history, body size or energetic state. Furthermore, individual variation in gastric evacuation rate was not correlated with feeding responsiveness at any temperature. These results indicate a general plasticity in the behavioural but not physiological aspects of energy acquisition.     

Early development and growth of the laboratory reared north-east Atlantic mackerel Scomber scombrus L.

The early development, growth and morphological changes of mackerel Scomber scombrus were investigated at different incubation temperatures (8, 10, 13, 15 and 18° C). Details on the early life history are illustrated with special reference to morphological transformations. Culture techniques to rear larval mackerel stages are described using laboratory cultured foods. Artificially fertilized eggs were hatched after 80·6 h at 18·4° C and 256·8 h at 8·7° C. The standard length ( L _S) of the individuals at first feeding was 4·71 ± 0·18 mm. Four mortality critical periods and cannibalistic behaviour were identified. A maximum average larval size of 37·5 ± 4·41 mm L _S was attained 30 days post-hatch (dph) at 18·4° C. Development and growth were affected significantly by temperature during both endogenous and exogenous feeding periods. Larvae grew more rapidly at high, than at low temperatures. Daily specific growth rate (in mass) ranged from 2·4% at 10·6° C to 16·9% at 18·4° C. Likewise, average growth rate (in length) ranged from 0·05 mm day⁻¹ at 8·4° C to 0·37 mm day⁻¹ at 18·4° C. The allometric relationship of L _S, with several body measurements was not affected by temperature. Comparison with larvae collected in the Bay of Biscay did not show any significant difference in the dry mass and L _S relationship; conversely, the growth rate in length differed significantly between both laboratory and field conditions. The trends observed in the laboratory are described in relation to some aspects of the year-class strength regulation.     

Observations on growth rates of Arctic charr, Salvelinus alpinus (L.), reared at low temperature

Food intake and growth of Arctic charr, Salvelinus alpinus , in fresh water was studied at three temperatures 2·9, 8·4 and 13·1° C). Best growth and highest food intake occurred at 13·1°C. The approximate chemical composition was dependent upon rearing temperature, fish reared at the highest temperature depositing large quantities of fat. Fish were later grown on in either fresh or salt water where two growth patterns were observed. In the light of published data for Salvelinus spp., it is suggested that the poorest growth was shown by fish which were incapable of complete adaptation to saltwater conditions.     

Laboratory studies on the effects of thermal change on the behaviour and distribution of juvenile chum salmon in sea water

Schooling chum salmon Oncorhynchus keta were biased towards the water surface (median position <1 m) under isothermal conditions (10° C) in a water column simulator (WCS). Thermal stratification (24/10° C) inhibited upward movement with fish congregating at the thermocline and displaying a clear avoidance of potentially lethal surface waters. A tri-phase model based on piece-wise nonlinear regression was used to describe the distribution shifts of chum salmon during a change from isothermal to thermally stratified conditions. Fish distribution was consistent with thermoregulatory behaviour and exhibited 'attraction', 'preference' and 'avoidance' phases. The thermal preference of 50% of the fish lay between 12·2 and 20·2° C, however, >83·5% of the fish occupied a 'preferred' temperature range of 13·7–17·9° C. The mean temperature at which 50% of chum salmon avoided rising temperature by shifting deeper in the water column and using the cooler thermocline was 20·2° C, and 90% avoidance occurred at 22·9° C. Behavioural responses to thermal stratification were consistent amongst underyearling fish of differing size and age.