Through the experiments presented here we wanted to test whether egg production of theblack-chinned tilapia Sarotherodon melanotheron heudelotii underexperimental conditions varies as a function of ambient salinity (fresh waters vs.sea waters vs. hypersaline waters, 0, 35 and 70, respectively)and whether these responses differ between fish acclimated within a few weeks from freshwater to saline and hypersaline environments (experiments E1 and E2, monitoring over 10and 18 weeks), and individuals born and raised all life long at the experimentalsalinities (E3, monitoring over 18 weeks). In total, 233 spawns were collected. In each ofthe three experiments, the reproductive investment (gram of egg per gram of female over 2weeks) did not differ between salinities of 0 and 35, whereas it was 2–3 times lower at 70than at 0–35, because of lower spawning frequency (E1-E3), smaller clutch size (E1) andlower spawn mass (E1-E3). Finally, fish acclimated to salinity from fresh water over a fewweeks and those maintained at a particular salinity all life long showed similarreproductive traits, thereby emphasizing the remarkable physiological plasticity of thisspecies.

This study relied on the day-by-day analysis of bioenergetics and prey size preference inisolated cannibals of the African catfish Heterobranchus longifilis(13–57 mm standard length, 3–500 mg dry body mass, n = 153) that were offeredad libitum conspecific prey of adequate sizes in small-sized (2-L)environments under controlled conditions (12L:12D, 30 °C). In these conditions, cannibals ofincreasing body size selected preferentially prey of decreasing size relative to theirown, and increasingly closer to the optimal prey size (producing the highest grossconversion efficiency). The role of experience in cannibalism was found of secondaryimportance relative to body size, both as regards food intake and prey size selectivity.These results indicate that in environments that minimize the escape capacities of theirpotential victims, as applies to most aquacultural contexts, fish exercising cannibalismtend to forage optimally, which has rarely been evidenced for piscivorous behaviour. Thepresent study further highlights that H. longifilis possesses a very highcapacity for growth, which originates from the combination of high food intake and veryhigh conversion efficiency, and makes this species of utmost interest for aquaculturistswherever fast growth is desirable, but also extremely prone to cannibalism whereverfeeding schedules are inadequate.

Hatchlings of clown loach (3.8 mm in total length, TL) are pelagic. When reared at 26−28 °C, they become benthic 3 days after hatching (dah), when their swim bladder is inflated. The fin development sequence (pectorals  <  caudal  <  dorsal  =  anal  <  pelvic) is typical of the clade Ostariophysi. All fins and finrays are fully elongated at 20 mm TL, but the finfold persists until 25 − 26 mm TL (start of juvenile stage). Melanophores appear at 3 dah, they form a 5-bar pattern at 5 dah, then two bars (III and V) vanish progressively, producing at 26 mm TL the 3-bar pattern that is typical of adults. Throughout the ontogeny, the pigment pattern exhibits a structural regularity (bars spaced at regular intervals), which is interpreted in a functional perspective by reference to the maintaining of crypsis and signalling throughout. Exogenous feeding commences at 4 dah (5.5 mm TL). Food intake (FI) increases rapidly, from 6% wet body mass (WM) at 5.5 mm TL to  >20% WM in fish  > 7 mm TL. Gut evacuation rate (Rg) increases with increasing meal size and fish size, as a result of gut coiling (from 8 to 15 mm TL), and is highest at 11 mm TL (about 10% WM h-1 in fish feeding maximally). The allometric increase of FI and Rg during the early larval stages is accompanied by increasing capacities for growth, so early sizes differences amplify rapidly during the ontogeny. Nevertheless, growth remains slow (mean of 0.4 mm TL day-1 from 4 to 29 dah; 0.9 mm TL day-1 for top growers). By contrast, unfed fish display long resistance to starvation (until 14−15 dah). The combination of slow growth and long resistance to starvation is discussed in respect to the reproductive phenology of the species, as the capacity of making metabolic economies prevails over fast growth for seasonal strategists spawning mainly at the start of the rainy season.

The knowledge of how fish survive and grow at different temperatures, and how these traits vary between life stages, is essential to evaluate the effects of climate change on wild fish and implement effective strategies in aquaculture. These issues are addressed in this study through a series of experiments that evaluate the effect of temperature (23–34 °C) on the embryos and larvae of clown loach, Chromobotia macracanthus. This species is endemic to the rivers of Sumatra and Borneo, highly praised on the ornamental fish market, and has been reproduced in captivity recently. No embryo survived a 24-h exposure to 34 °C until the age of 3 days after hatching (dah); mortality was high at 32 °C at 2 and 3 dah, whereas it was low and similar from 1 to 4 dah at 23–29 °C (<10%). Yolk absorption was proportional to water temperature (Q10°C of 1.69 in the 23–32 °C range), but fish reared at cold temperatures were larger than others at the start of exogenous feeding (5.7 vs. 5.5 mm TL, at 23 and 32 °C, respectively). The survival of larvae fed Artemia nauplii ad libitum was high at 23–32 °C (80–100%), but almost null at 34 °C. Growth models at different temperatures were produced from weekly measurements in two experiments, and tested by comparing their predictions with the results of a third experiment. Throughout the larval stage, the optimal temperature for growth (T°opt) was close to 29 °C, and departures from T°opt resulted in substantial growth penalties (–30% SGR for –5.1 °C and  + 3.1 °C). High survival, fast growth (0.7 mm day-1) and limited size dispersal at T°opt are encouraging perspectives for the aquaculture of clown loach. From an ecological perspective, the species has an atypical thermal biology, as it is less thermophilic than other tropical fishes, but more stenothermal than temperate fishes exhibiting similar values of T°opt , both traits being of particular concern in the context of global warming.

In comparison to older life stages, the embryonic stages of fishes generally have narrow tolerance ranges for environmental conditions, as regards water quality, temperature and mechanical shocks. The knowledge of these factors is indispensable to appraise the threats brought about by climate or anthropogenic changes upon their resilience, and to define adequate ways of incubating their eggs for an efficient propagation of the species under controlled conditions. Clown loach eggs have a narrow thermal tolerance range in comparison to other tropical and temperate fishes. Hatching occurs at 22–30 °C, and non-deformed larvae can only be obtained at 23.8–30.2 °C. Furthermore, the thermal tolerance of any particular progeny was found dependent on the maintenance temperature of the female parent, thereby making the actual tolerance no broader than 4.5 °C. The (log-log) relationship between the duration of the incubation period and temperature was characterized by a shallow slope, which is more typical of coldwater fishes, as is a narrow thermal tolerance range. On the other hand, clown loach hatched more rapidly (20 h at 26 °C) than predicted by existing models on the basis of water temperature and egg diameter, a feature that is shared by other warmwater fishes producing eggs that undergo a strong swelling process (about three times the ova diameter in clown loach). Clown loach embryos are strongly sensitive to mechanical shocks, but their development is not viable either in protracted steady state conditions, in absence of water movement, as they develop various deformities (e.g. pericardial oedema). This is thought to originate from a hypoxic microenvironment around the embryo, as a consequence of an oxygen gradient developing inside and outside the egg, since the boundary diffusion layer is not refreshed by water movement. This issue is worsened by strong egg swelling and incubation at warm temperature.

The clown loach Chromobotia macracanthus, endemic to Indonesia, is a major species on the international market of ornamental freshwater fish. In order to satisfy an increasing demand with a sustainable alternative to the massive capture of wild juveniles, research has been dedicated to the artificial propagation and domestication of this species. The present study, the first of a series, focused on favourable maintenance conditions for broodfish sexual maturation, criteria for identification of ripe fish, efficiency of hormone-induced breeding treatments, predictability of their latency response, and on the comparison of reproductive performances of fish from populations of Sumatra and Borneo Islands (in total, 112 females of 46 to 404 g body weight). When reared in fully controlled conditions in large water recirculation systems, broodfish originating from Sumatra had reproductive performances similar to or slightly higher than those maturing in the wild (ovulation rate of 93% vs. 82%, relative fecundity of 109 277 vs. 103 550 ova  kg-1 and fertilization rate of 73% vs. 61%, respectively). In the same rearing conditions, captive females from Borneo (n = 22) showed lower ovulation rate (77%), relative fecundity (76 262 ova kg-1) and fertilization rate (50%) than those originating from Sumatra (n = 28). By contrast, the mean individual weight of ova (around 0.8 mg) was independent from the origin or maintenance conditions of females. An initial modal follicle diameter  ≥1.02 mm generally led to high ovulation success (>80%) after hormonal treatment and is recommended as the main criterion for selecting female broodfish. Two hormonal treatments for inducing oocyte maturation and ovulation (T1: two successive injections of Ovaprim at a 6 h-interval; T2: one injection of human chorionic gonadotropin (hCG)- and one of Ovaprim 24 h later), produced similar results in terms of ovulation rate, quantity and quality of ova collected. With both treatments, the latency decreased with increasing water temperature, then increased again at temperatures  >28–29 °C. To our knowledge, such U-shaped relationship between the latency response and temperature has never been documented in teleost fishes.

Meal frequency is a key parameter in fish larviculture, especially in highlycannibalistic species. Knowledge of the biological bases of cannibalism (growth capacityof cannibals, morphological constraints on cannibalism, prey size preference) can helppredicting the risks of cannibalism for different feeding schedules under cultureconditions. This study relied on the day-by-day analysis of prey size preference andbioenergetics of individual cannibals of the catfish Pseudoplatystoma punctifer(8–65 mm standard length, SL, 0.5–400 mg dry mass,DM) at 28.5 °C under 12L:12D. The results were equated with theontogenetic variations of morphological factors (head and mouth width) and feed efficiencyof larvae feeding on Artemia nauplii, in order to calculate the risks ofcannibalism among fish fed 2–7 daily meals. The predation capacities of P.punctifer were highest at 8 mm SL and decreased in larger fish(largest prey  =  86% and 70% SL in fish of 8 and  >30 mmSL, respectively). Cannibals of increasing size preferred increasinglysmaller prey relative to their own size, but also to their predation capacities. Thesemorphological and behavioural constraints were largely compensated for by bioenergeticsperformance. Cannibals consumed high daily food rations (as high as 171 and 29%DM in fish  <1 and  >300 mg DM,respectively), exhibited high gross conversion efficiencies (0.50–0.55 and about 0.70, infish  <1 and  >30 mg DM, respectively), and grew rapidly(90 and 18% DM day-1 in fish  <1 and  >300 mgDM, respectively). The growth advantage of cannibals over siblings fedArtemia nauplii was decisive, except for high meal frequencies (6–7daily meals). This study supports the view that the risk of cannibalism and adequatefeeding strategies can be largely predicted in a particular fish species if themorphological, behavioural and bioenergetics bases of cannibalism are examined altogetherin an ontogenetic perspective.

Because preying on close relatives may cause a loss of inclusive fitness, cannibalistic animals are generally deemed to cannibalise non-related conspecifics preferentially to kins. Whether this basic principle also applies to fish exerting intracohort cannibalism at an early developmental stage is uncertain, and more intense cannibalism among mixed progenies might just be a consequence of greater size heterogeneity. To investigate this, progenies of sharptooth catfish of equivalent initial body weights but with contrasting size heterogeneity were reared separately or in mixed groups containing equal proportions of each progeny (27 ± 1°C, 12 h light:12 h night, 2.5 L aquaria, 100 or 200 fish·L–1). Cannibalism was monitored on a daily basis until the end of the larval stage (3–15 days after hatching). Cannibalism in mixed groups was intermediate between those in pure progenies, and was positively correlated (R2 = 0.803, P < 0.01) with initial size heterogeneity, irrespectively of fish origin. This finding does not exclude that catfish larvae were able to discriminate between siblings and non-related fish, but this obviously had very little influence on cannibalism. The implications of this finding are discussed, as they apply to prey selection in fast growing larvae, and aquaculture of catfish.

Radio tracking has been extensively used to enhance our knowledge on the movement and home range of fish in general and salmonids in particular. However, the use of various temporal sampling protocols is likely to overlook fish movements, and produce experimental artefacts, the amplitude of which is unknown and may compromise comparison of fish behaviours revealed by different protocols. Starting from a day-by-day tracking study of brown trout in Belgian waters (Aisne stream, nine fish, minimum 39 daily locations per fish), we re-examined their home range and mobility, through a subsampling process, as if the fish had been located at longer time intervals (2–14 days). The estimates from subsamples were compared to the original data set in order to quantify the corresponding reduction of accuracy from observed data, and how far this could be predicted on the basis of locating frequency. The results clearly indicate that all intervals longer than one day generally generate substantial biases (reduction of accuracy from 0 to 82% for home range and from 5 to 92% for mobility) but these can be partly corrected through the use of predictive models. This analysis demonstrates that any comparison between studies relying on different locating frequencies can generate some ambiguity when interpreting biological phenomena or geographical differences.

Pirapatinga given submaximal rations grow faster when fed at night than when fed during daylight hours. It was hypothesised that these fish were day–night conformers, with a nocturnal acrophase, and expended more energy during daytime feeding. In order to test this hypothesis, telemetry was applied to study the behaviour of cultured pirapatinga (stocking biomass of 40 kg·m–3; 27 ± 1 °C; 6 mg O2·L–1; 13.5 L/10.5 D) under three feeding schedules (diurnal versus nocturnal feeding [12 g·d–1·fish–1 over 10 h] versus fasting) that were evaluated in succession over 1 week each. This implied a feasibility study to test the adequacy of tagging fish with surgically implanted transmitters (ventrolateral incision, posterior to the pelvic girdle). There was no mortality or tag loss over 12 months following surgery, and abdominal incisions healed within 4 weeks. Under all three feeding schedules, activity increased at dawn, peaked during daytime, decreased at twilight and reached a minimum during the late night. Diurnal and twilight activity levels were similar under all three schedules, whereas nocturnal activity was significantly lower with daytime feeding. These observations indicate that pirapatinga have a diurnal activity acrophase, which is little influenced by food availability, and imply that the higher growth of fish fed at night does not originate from lower energetic expenditures.

Sibling juvenile vundu Heterobranchus longifilis reared at 27 ± 1 °C under 12L:12D were fed during the day, at night or over the entire 24-h cycle, for periods of 14-15 d. Nighttime feeding gave the highest growth rates, best feed conversion, lowest mortality, fewest losses due to cannibalism and least growth heterogeneity. The trends were seen in fish of all ages and weights (31−101 d and 0.3−30.0 g), but differences were significant only for the smallest fish that were reared at the lowest stocking biomass (< 3.5 g·L−1). The effects of daytime and nighttime feeding were examined in 151-d old (85 g) juveniles stocked at different biomasses (3.4 and 13.6 g·L−1). The growth of vundu was density-dependent, being impaired at low stocking density, but this was substantially mitigated under nighttime feeding. Rearing at high stocking density, feeding at night or in darkness, may act through similar mechanisms that involve changes in the level of aggressive behaviour. It is suggested that nighttime feeding in intensive catfish culture could be replaced by 24-h feeding under conditions of permanent darkness.

A biotelemetry experiment was designed to determine whether the selection of a residence area by adult barbel Barbus barbus (Cyprinidae) was dependent on the availability of physical habitat features, on the presence of fish shoals and/or on the presence of specific partners. The experiment took place in 3.0 km stretch of the River Ourthe (Belgium), where the latter features had been characterised by habitat and electric fishing surveys (1 190 fish ≥ 15  cm, at 265 locations). Six barbel (34.8–49.5 cm) belonging to a single shoal of 34 fish were tagged with 40 MHz radio transmitters, transplanted in pairs at different distances (−1 km, −30 m, +1 km) from the capture site and tracked from 3 September to 7 October 1990.

Fish showed variable mobility patterns and home range sizes (200–2 400 m). They selected 17 residence areas of which the cumulative frequency of occupation ranged from 0.1 to 79.5 day × fish. The comparison between the residence areas selected, habitat suitability and the location of fish shoals in the study area indicate that the presence of a shoal of at least 10 resident fish within a suitable habitat (normalised suitability index ≥ 1.0) is necessary but not sufficient to guarantee its long term selection and occupation. Homing movements of transplanted fish were directed towards the activity area rather than to the residence area itself. The fish that did not home, selected residence areas in a part of the river with a higher availability of potential feeding areas. These elements are discussed within the context of habitat utilisation and foraging strategies by a gregarious species. It is suggested that resident fish of the same species may ease the exploitation of a non familiar environment by a naive fish, and that this may positively trade off the benefit provided by home site fidelity, depending on the availability of food resources.

Surgery procedures were used to implant telemetry transmitters into the body cavity of adult (574–1 033 g) tilapias Oreochromis aureus in aquaculture tanks (4 m2, 1.5 m3, 26.5 ± 0.5 °C, ≥ 5.0 mg O2 l−1) and their effect on fish survival, growth and behaviour was evaluated. Only one out of 35 implanted fish died. With one exception, all fish consistently retained their transmitter until the end of the study (up to 30 months). Healing was faster when the incision was sutured with polyamide monofilament (5–14 days) than with other suture materials, due to tunnelling with atraumatic needles for catgut or fouling of braided silk. In all 10 fish sacrificed after 30 and 50 days, the transmitter had become encapsulated by connective tissue. No infection or damage to the viscera was observed.

The activity of four tilapias (903–1 033 g) equipped with motion sensitive transmitters was telemetered during the recovery from anaesthesia and surgical procedures. All four fish maintained a normal diurnal activity rhythm pattern throughout the study but had low levels of activity during the first 12–24 h. Based on the evolution of their resting posture after surgery, it is suggested that tilapias need 3 to 4 days to completely compensate the negative buoyancy resulting from anaesthesia and tagging.

L'étude du régime alimentaire de C. nigrodigitatus Lacépède (N = 646$) et C. auratus Geoffroy Saint-Hilaire (N = 253) a été réalisée d'après l'analyse des contenus stomacaux de poissons du lac Nokoué et de la lagune de Porto-Novo (Sud-Bénin) en 1990-1991. L'analyse factorielle de correspondance sur l'occurrence de 16 catégories de proies dans 21 catégories de poissons (suivant l'espèce, la taille des poissons, la saison et l'habitat) met en évidence la répartition des ressources alimentaires entre C. auratus et différentes catégories de tailles de C. nigrodigitatus capturés hors acadjas (enclos en branchages). Les indices d'abondance (Iab) et de volume (Iv) des proies et leur régularité (Rab et Rv) dans le régime alimentaire mettent en évidence la tendance benthophage des deux espèces : C. auratus (6-20 cm Ls) est beaucoup plus généraliste (Rv = 0,842) se nourrissant essentiellement de mollusques et de petits crustacés (branchiopodes, copépodes, ostracodes) par fouille du substrat. C. nigrodigitatus, en revanche, se spécialise avec l'âge et la taille (Rv = 0,826 à 0,597 de 6-10 cm à 6-20 cm Ls) vis-à-vis des décapodes (Iv de 13,4 à 49,4 %) et des poissons (œufs et alevins, Iv de 7,8 à 66,9 %). Les différences saisonnières concernent principalement la période des crues, au cours de laquelle les poissons se nourrissent de larves ou de nymphes d'insectes sur des végétaux de la plaine inondée. Dans les acadjas, les deux espèces partagent les mêmes ressources alimentaires. Cette étude de l'écologie trophique des Chrysichthys du complexe lagunaire du Sud-Bénin, mise en parallèle avec leurs potentialités de croissance, suggère que le régime alimentaire généraliste de C. auratus se situe dans un contexte de ségrégation sélective vis-à-vis de C. nigrodigitatus. La valeur adaptative des deux espèces dans un écosystème modifié et exploité par l'homme est brièvement discutée.

This paper presents a comprehensive study of the impact of damming on the spawning migrations of Barbus barbus in the canalized River Meuse (Southern Belgium). A Denil fish pass on the Ampsin-Neuville dam was controlled 251 times in 1989-1993. The most striking feature is the almost complete absence of barbel in 1990-1993 (15 fishes in 4 years). Most captures of barbel in the fish pass in 1989 were clumped within a few days (mid-May) and related with spawning migrations. Stepwise multiple regression analyses revealed that attractivity was the major condition set allowing barbel to migrate successfully through the pass, while feasibility parameters (water velocity) were found not to be relevant. The variables involved in the attractivity condition set refer indirectly to the influence of water catchment by a hydroelectric plant and to the relative importance of the flow in the pass. The study concludes that this additional condition set significantly interferes with the natural environmental stimuli triggering spawning migrations in barbel and questions the effectiveness of the thermally-related reproductive strategy of the species in an environment with restricted longitudinal connectivity. This statement is discussed in parallel with the recent evolution of barbel populations in canalized rivers.

Two female barbel Rarbus barbus (L.) were tagged with implantable radio transmitters and tracked in the Ourthe River (Southern Belgium) from April 12th to July 7th, 1989 to study their movements and activity rhythms during the spring-summer period. Both barbels defined a 1.6 km home range and occupied four residence areas. Greatest average daily movements within the home range limits occurred during the prespawning and spawning periods and were associated with variations of mean daily water temperature and level. Barbel 2 made a 9.6 km excursion from its spring home range and homed a week later in its residence area. Barbels displayed dusk and dawn activity patterns, character- ized by the occupation of riffle areas. The activity timing was found to be dependent on water temperature while the size of daily activity area was more influenced by river flow conditions.