Reviewers: N/A
Editor: Billo Heinzpeter Studer

• Initial release: 2019-05-09
• Appearance version: 2022-06-12
• Major version: 2019-05-09
• Minor version: 2022-01-22

Cite as: »Volstorf, Jenny. 2022. Clarias gariepinus (Dossier). In: fair-fish database, ed. fair-fish. World Wide Web electronic publication. Version B | 1.1. https://fair-fish-database.net.«

• Unpredictable influence:
  • WILD: caught individual had preyed on frog Leptodactylus ocellatus threatening to heavily influence the native food web: Guaraguacu river, Parana, Brazil (introduced) 1.
  • WILD: individuals of up to 25 years of age indicate persistence in invasion of non-native habitats: Darling dam, South Africa (introduced) 2.
• Competition:
  • WILD: local fisherman reported increase in catching Clarias gariepinus while decline in catching native species: Guaraguacu river, Parana, Brazil (introduced) 3.
  • WILD: 1 x 1 x 0.5 m exclusion cages with mixture of stones and polythene strips were placed in river at depth of 0.5-0.8 m on pebble and silt bottom for four weeks. Three weeks "before" phase: Lower number of macroinvertebrate taxa in cages from river which C. gariepinus had invaded than in uninvaded and fishless river (14 versus 22). Mainly leeches Hirudinea in invaded river, mainly oligochaete earthworms in uninvaded river.
    Pair of individuals (20-40 cm) introduced in cages ("treatment"). During three weeks "after" phase: Increase in diversity of macroinvertebrates and taxa richness in control cages in invaded river, no change in cages with individuals. Decrease in diversity in treatment cages in uninvaded river compared to control. Increase in mean dry mass in control cages in both rivers and increase in treatment cages in invaded, decrease in treatment cages in uninvaded river. Results indicate depletion of macroinvertebrates in uninvaded river and probable adaptation to invader in previously invaded river (e.g., dispersal, hiding): tributaries of Great Fish river, South Africa 4.
  • WILD: caught individuals had preyed on rare and endangered species of the region: cichlids, carp, eel, macropods, anabandits. Local fisherman reported increase in catching C. gariepinus while decline in catching native species. Competition for food with native conspecifics Clarias fuscus and Pelteobagrus fulvidraco in the same habitat: Pearl river, South China (introduced) 5.
  • WILD: low diet overlap between C. gariepinus and Centropomus prallelus indicating food partitioning: Itanhem river estuary, Bahia, Brazil (introduced) 6.
• Disease transmission: no data found yet.
• Interbreeding: no data found yet.

• For feeding → 7 8.
• For daily rhythm → 9 7.
• For swimming → 9 7 4.
• For migration → 10 11 7 12.
• For courtship, spawning → 11.
• For electrical communication → 13.
• For aggregating → 7.
• For social behaviour → 11.
• For cooperation → 7.

• For feeding → 11 14 15 16.
• For daily rhythm → 11 17 18 19.
• For swimming → 20.
• For touching → 19.
• For electrical communication → 21.
• For dominance and subordination → 21 19 16 22.
• For cannibalism → 11 14 23 24 25.
• For cooperation → 7.
• For aggression → 21 26 27 19 16.
• For territoriality → 17.
• For learning → 27 18.
• For coping styles → 19 28 29 16.
• For observable stress reactions → 14 17 26 27 30 31 32 33 34 29 25 35 36.

• Observations Africa: Agbokim waterfalls, Cross River State, Nigeria 12, lake Awassa, Ethiopia 8, lake Chamo, Ethiopia 13 38, lake Edward on border between Democratic Republic of the Congo and Uganda 39, lake Ngezi, Zimbabwe 7, lake Sibaya, South Africa 40, Nigeria 41, Okavango delta, Botswana 42, P.K. le Roux dam, South Africa 43, river Shire, Malawi 10.
• Observations Israel: Hula nature reserve, Israel 44.

• Observations Africa: Darling dam, South Africa 2, Great Fish and Sundays rivers, Glen Melville dam, South Africa 4.
• Observations Asia: Hanjiang and Pearl river basin, South China 5, Vembanad lake, Kerala, India 45.
• Observations South America: Doce and Grande river basin, Minas Gerais, Brazil 46, Guaraguacu river, Parana, Brazil 3, Itanhem river estuary, Bahia, Brazil 6, lagoa Encantada, Bahia, Brazil 47, Sao Francisco river basin, Minas Gerais, Brazil 46.

• Observations Banded tilapia WILD: Tilapia sparrmanii: lake Sibaya, South Africa 9.
• Observations Bluefin dwarf WILD: Centropygre multispinis: lake Sibaya, South Africa 9.
• Observations Burrowing goby WILD: Croilia mossambica: lake Sibaya, South Africa 9.
• Observations Eastern bottlenose elephant snout WILD: Mormyrus caschive: lake Chamo, Ethiopia 13.
• Observations Estuarine round herring WILD: Gilchristella aestuaria: lake Sibaya, South Africa 9.
• Observations Leaden labeo WILD: Labeo molybdinus: lake Sibaya, South Africa 9.
• Observations Natal topminnow WILD: Aplocheilichthys myaposae: lake Sibaya, South Africa 9.
• Observations Redbreast tilapia WILD: Tilapia rendallii: lake Sibaya, South Africa 9.
• Observations Sibayi goby WILD: Silhouetta sibayi: lake Sibaya, South Africa 9.
• Observations Southern mouth-brooder WILD: Pseudocrenilabrus philander: lake Sibaya, South Africa 9.
• Observations Striped topminnow WILD: Aplocheilichthys katangae: lake Sibaya, South Africa 9.
• Observations Tank goby WILD: Glossogobius giuris: lake Sibaya, South Africa 9.
• Observations Wahrindi WILD: Synodontis schall: lake Chamo, Ethiopia 13.

• Plants:
  • WILD: inundated grassland and forest fringes including Phragmites mauritianus Kunth, Scirpus littoralis Schrad, Andropogon amplectens Nees, Imperata cylindrica (L.) Beauv., Cladium spp., Juncus kraussii Hochst., Panicum meyerianum Nees., Eragrostis capensis (Thunb.) Trin., Harpochloa spp., in terrace habitats also submerged macrophytes including Potamogeton pectinatus L., P. schweinfurthii A. Benn, Myriophyllum spicatum L., in gradual sloping habitats also Typha latifolia L. and beds of submerged Ceratophyllum demersum, in sheltered bays and inlets also Nymphaea capensis Thunb.: lake Sibaya, South Africa 9.
  • WILD: grasses Echinochloa along the banks, in lagoons dense beds of Ceratophyllum, Nymphaea, Typha: Elephant marsh of river Shire, Malawi 10.
  • WILD: individuals were found over drowned trees: Glen Melville dam, South Africa (introduced) 4.
  • For substrate and spawning → D1.
• Rocks and stones: no data found yet.
• Sand and mud:
  • WILD: individuals were found over mud and shale: Glen Melville dam, South Africa (introduced) 4.
  • WILD: JUVENILES-ADULTS were found over mud: Shesher and Welala wetlands, lake Tana, Ethiopia 48.
  • WILD: individuals (685-1,240 mm) were found over mud and shale: Darling dam, South Africa (introduced) 2.
  • WILD: ADULTS were found over terrace habitats with sandy and detritus bottom: lake Sibaya, South Africa 9.
  • LAB: 7 day-old FRY in 250 m³ aquarium with Myriophyllum spicatum and sand foraged in sand and ejected grains through operculum 11.
• Other substrate: no data found yet.

• FARM: eggs were collected from mature females injected with suspended pituitary and mixed with milt. Eggs were incubated in conrete flow-through troughs either on roots of Nile cabbage (Pistia stratiotes), water hyacinth (Eichhornia crassipes), pond weed (Ceratophyllum dermasum) or leaves of green grass (Commelina sp.) or unnatural substrates: 1,350 cm² mats of kakaban, sisal, papyrus or nylon. Highest hatching rate compared to control (concrete slabs) on Pistia stratiotes (66.2% versus 18.6%), followed by Commelina sp. and Eichhornia crassipes (54% and 49.7%). Hatching rate on Ceratophyllum dermasum lower than control (13% versus 18.6%). Among unnatural substrates, no mat exceeding hatching rate on control substrate. No difference in sisal mat compared to control (18.6%); papyrus (12%), kakaban (11.8%), and nylon (4%) lower than control 49.
• FARM: JUVENILES in earthen ponds fertilised with pig manure and installed with bamboo poles (diameter 9 cm, length 90 cm) at density of 4 poles/m². After 90 days, higher gross yield (16.0 kg/100 m² versus 6.0 control versus 9.5 additional feed) and higher specific growth rate (1.2%/d versus 0.5%/d control versus 1.1%/d additional feed) compared to control and compared to condition in which individuals were additionally fed three parts groundnut husk, one part wheat bran at 5% per pond biomass. Results indicate advantage of bamboo poles probably by enabling periphyton growth that served as food 50.

• Plants:
  • WILD: JUVENILES-ADULTS were found preferably under cover of vegetation: Elephant marsh of river Shire, Malawi 10.
  • WILD: individuals (59.5-70 cm, 2.3-3.7 kg) were observed in social aggregations under growths of algae at night: lake Ngezi, Zimbabwe 7.
  • LAB: 2 day-old FRY in 250 m³ aquarium with Myriophyllum spicatum and sand stayed close to vegetation during the day, explored aquarium at night 11.
• Rocks and stones:
  • WILD: individuals (59.5-70 cm, 2.3-3.7 kg) were observed in social aggregations under rocks at night: lake Ngezi, Zimbabwe 7.
• Sand and mud:
  • WILD: JUVENILES-ADULTS were found occasionally under shelter of mud banks: Elephant marsh of river Shire, Malawi 10.
• Other cover: no data found yet.

• LAB: FRY in 30 L glass aquaria longitudinally divided by glass partition, stocked with either 50 or 150 FRY and either equipped with black plastic shade material (500 x 250 mm) with mesh size 8 x 4 mm or left barren. At first observation on day 6, higher cannibalism rate in compartments without shelter than with shelter (ca 1.7-1.9% versus 0-0.8%). Missing or minimal cannibalism under condition with shelter and 50 IND/tank until second observation on day 12. Difference between conditions decreased with increasing time to about 0.5% in all densities at end of observation period on day 47. In general, among 50 interactions, higher number of territorial aggressive acts when provided shelter than without shelter (50 IND/tank: ca 32 versus 18 acts, 150 IND/tank: ca 13 versus 4 acts), but overall, lower browsing (ca 20-21% versus 66-72% across densities) and higher resting activity when provided shelter than without (ca 77-78% versus 26-32% across densities) and consequently fewer incidences of contact 14.
• For shelter and survival under different stocking densities → D2.

• LAB: FRY in 60 L white plastic bins at either 0 or 24 h light (250 lux) and either cover by black nylon shadecloth netting or not. After 13 days, higher growth rate in 0 h light than 24 h light (1.4 mm/d versus 1.1-1.3 mm/d). No difference between cover or no cover in 0 h light, but higher growth rate in 24 h light when provided cover (1.3 versus 1.1 mm/d). No difference in condition factor (1.1) and mortality (14-20%) 17.
• For cover and growth and tank colour → D3.

• Observations: 3.8±0.4 se 51.

• Carnivorous to omnivorous D4. The fishery that provides fish meal and fish oil has two major impacts:
  1. It contributes considerably to overfishing, as it accounts for 1/4 52 or even 1/3 53 of the world catch volume.
  2. It challenges animal welfare, because in the face of 450-1,000 MILLIARD wild fishes caught worldwide each year to be processed into fish meal or fish oil 54, the individual fish gets overlooked and, thus, suffering increases at rearing, live marketing, and slaughtering levels 55.

• Food items: carnivorous, omnivorous if need be:
  • Observations WILD, JUVENILES-ADULTS: humus and plant detritus, fish (Cichlids, Barbus spp., Clarias), insect larvae (e.g., Chironomids, Dragonfly nymphs), filamentous algae: Elephant marsh of river Shire, Malawi 10, Copepods (mainly Thermocyclops, then Mesocyclops) highest relative importance (80.5% and 11%), followed by Cladocera (2.9% Moina, 0.8% Ceriodaphnia, 0.7% Diaphanosoma), detritus (3.1%), fish (0.5%), Rotifera (0.3% Brachionus), fish scales (0.2%), insects (0.07% Chironomidae larvae, 0.01% Anisoptera): lake Chamo, Ethiopia 38, mainly fish (e.g., goldie barb Barbus pallidus, Mozambique tilapia Oreochromis mossambicus, moggel Labeo umbratus), aquatic invertebrates (e.g., Ephemeroptera, Trichoptera, Hemiptera, Odonata, Diptera, Lepidoptera, Decapoda, Mollusca), Macrophytes, terrestrial insects, detritus: Great Fish and Sundays rivers, Glen Melville dam, South Africa (introduced) 4, in wet season (May-November): mainly fish (juveniles, fish remains, whole adults), to a lesser extent: insects (adults, stages), crustaceans (Decapods, bivalves, Gastropods), rotifers; in dry season (December-April): crustaceans (Decapods, Gastropods, bivalves), insects, plant material, fish: Agbokim waterfalls,
    Cross River State, Nigeria 12, mainly shrimp Caridae, vegetable organic matter, insects, Gastropoda: Itanhem river estuary, Bahia, Brazil (introduced) 6.
  • For cannibalism → D5.
• Food items and habitat: no data found yet.
• Food items and life stages: mainly aquatic invertebrates, increasing proportion of fish with increasing age, occasional cannibalism:
  • WILD: tendency of decrease in insect larvae, increase in fish, but small sample size of large ADULTS: Elephant marsh of river Shire, Malawi 10.
  • WILD: lake Awassa, Ethiopia 8:
    JUVENILES: mainly fish, especially O. niloticus, of lesser importance Aplocheilichthyes, further: macrophytes and roots, insect larvae (Anisopteran, Zygopteran, Chironomidae),
    ADULTS: mainly fish, especially O. niloticus (individual of 52.5 cm and 1,100 g had fed 23.5 cm, 210 g O. niloticus), of lesser importance Garra sp., Barbus sp., occasionally C. gariepinus, further: macrophytes and roots, detritus.
  • WILD: lake Chamo, Ethiopia 38:
    highest relative importance in diet of JUVENILES (17.1-44.8 cm TOTAL LENGTH): Copepods (59.8% Thermocyclops, 14.1% Mesocyclops), followed by detritus (18.1%) and Cladocera (3.5% Moina, 1.7% Ceriodaphnia, 0.05% Diaphanosoma), fish (0.9%), fish scales (0.8%), insects (0.6% Chironomidae larvae, 0.02% Hemiptera, 0.01% Ephemeroptera), Rotifera (0.4% Brachionus);
    highest importance in diet of ADULTS (55.2-109 cm): Copepods (84.5% Thermocyclops, 10.1% Mesocyclops), followed by Cladocera (3.6% Moina, 0.8% Ceriodaphnia, 0.1% Diaphanosoma), fish (0.5%), detritus (0.3%), fish scales (0.1%).
  • WILD: individuals <25 cm mainly aquatic invertebrates, individuals from 25 and especially >50 cm mainly fish: Great Fish and Sundays rivers, Glen Melville dam, South Africa (introduced) 4.
• Food preference:
  • LAB: FRY in 30 L aquaria either fed a) live food (plankton), b) plankton and trout starter, or c) plankton and betaine-supplemented trout starter twice daily. After seven days, higher weight in groups with trout starter compared to plankton-only group (14.7-15.0 mm and 23.4-24.9 mg versus 8.8 mm and 4.5 mg) and highest survival in group with plankton and trout starter (66.4 versus 31.2-40.8%). In the latter group, highest preference for Rotifer Rotaria sp. (325 µm), followed by Cladocer Chydorus sphaericus (650 µm) and Copepod Megacyclops viridis (2,000 µm) 58.
  • For food preference and pellet colours → D6.
• Food partitioning: no data found yet.
• Prey density: no data found yet.
• Prey size selectivity:
  • LAB: FRY (15.8 mm) and JUVENILES (59.6 mm) in 30 L glass aquaria were presented with: 42% small ostracods (0.6 mm), 15% large ostracods (1.6 mm), 31.7% Daphnia (2.5 mm), 7.6% small mosquito larvae (2.7 mm), 1.2% medium mosquito lavae (6.0 mm), 2.5% large mosquito larvae (7.8 mm). FRY and JUVENILES preferred larger items (three types of mosquito larvae), although smaller items were more abundant 14.
  • LAB: prey-naive JUVENILES (mouth width 18 mm) in 100 L aquaria. During 10 days, introduced to various prey. Avoided Topmouth gudgeon (Pseudorasbora parva, height 7 mm) and Nile tilapia (Oreochromis niloticus, height 12 mm), preferred Rudd (Scardinius erythrophthalmus, height 9 mm) and Sunbleak (Leucaspius delineatus, heigth 8 mm) 15.
• Particle size:
  • WILD: prey was 1/8 to maximum 1/4 the size of the predator: Elephant marsh of river Shire, Malawi 10.

• WILD: JUVENILES caught dragonfly larvae rather during the day than the night (42.4% versus 11.4 frequency of occurence), O. niloticus fry rather during the night (77.3% versus 12.1% frequency of occurrence): lake Awassa, Ethiopia 8.
• WILD: individuals (59.5-70 cm, 2.3-3.7 kg) were observed pack hunting during the day by cruising an area in shoals: lake Ngezi, Zimbabwe 7.
• WILD, JUVENILES-ADULTS: large proportion of zooplankton in diet indicates ram feeding: lake Chamo, Ethiopia 38.
• LAB: FRY in 250 m³ aquarium with Myriophyllum spicatum and sand or in an enamel tray searched for food at 80 h after hatching on water surface and on substrate 11.
• LAB: FRY in 200 x 90 x 90 mm glass aquaria were presented with 50 large Daphnia. Searched for food, lunged or sucked when they came in contact with Daphnia, and captured them only then 14.
• LAB: FINGERLINGS grabbed pellets and swallowed or rejected it, rarely grabbed twice or even three times. Held pellets in the mouth for 1.8-4.7 s before swallowing (for details on the study → D7) 16.
• LAB: JUVENILES wounded prey and fed on it when dying. JUVENILES did not successfully capture selected prey even though JUVENILES were starving and height of prey was smaller than mouth width (→ D4). These selected prey were less susceptible to wounding. Results indicate that height of prey alone is not decisive for prey selectivity but rather susceptibility to wounding. Also, poor specific growth rate (0.4%/d) and FOOD CONVERSION RATIO (4.7) after 10 days exclusively feeding on live fish indicates loss of too much energy for prey capture 15.
• For feeding style and...
  ...(non-)importance of vision → D8,
  ...importance of mechanical sensing → D9.

• LAB: FRY in 12 L tanks (12 x 30 x 23 cm) at densities of 5, 13, 22, or 30 IND/L and either fed three (08:00, 12:00, 16:00 h) or six times a day (08:00, 10:00, 12:00, 14:00, 16:00, 18:00 h). After 30 days, higher aggression (20.7 versus 16.1 contacts/5 min) and higher number of total contacts (aggressive, non-aggressive, disturbed rest) under six times than three times feeding schedule (41.6 contacts/5 min versus 29.8 contacts/5 min). No effect of density. During 10 s of swimming (66% versus 56%) and during 10 s of browsing, more aggressive contacts at higher than lower feeding frequency (1.5 versus 0.9 contacts/10 s) 26.
• LAB: JUVENILES in 120 L aquaria (90 x 45 x 45 cm, water depth 30 cm) and either fed continuously by self-feeder or two times a day by hand. During six weeks, less swimming activity and more resting under self-feeding than under hand-feeding (145.9-221.5 versus 153.3-229.2 s swimming activity during 5 min; 78.5-154.1 versus 34.6-146.7 s resting during 5 min). Also, less JUVENILES bitten by conspecifics (39.6% versus 47.4%) 27.

• LAB: FRY in round 10 L tanks with diameter-to-depth-ratio of 10 and flow rate that did not require them to swim at 30 °C. Satiation time (the time until satiated) increased with increasing age, from 5 min at day 1 to ca 30 min from day 5 on. Stomach capacity until satiation increased linearly with increasing FRY weight, from ca 0.5 mg feed at ca 3 mg FRY weight to ca 11 mg feed at ca 58 mg FRY weight. After being fed decysted Artemia to satiation at day 6 or 7, stomach content represented ca 17% body weight. Decreased exponentially with increasing time without food: at 10 h since satiation, below 10% body weight; at 20 h at ca 2% body weight. FRY were offered Artemia again and fed increasingly the longer the deprivation time, e.g. ca 7.5% body weight after 2 h, ca 16% body weight after 16 h, ca 18% after 24 h. Weighing the benefits of amount of feed and feeding frequency, the authors propose different feeding schedules, e.g. every 4 h on a 12 h schedule with 41% of daily ration on first ration and 20% daily ration each on three subsequent rations (for more info → Table 8.3) 24.
• LAB: FRY in 12 L tanks (12 x 30 x 23 cm) at densities of 5, 13, 22, or 30 IND/L and either fed three (08:00, 12:00, 16:00 h) or six times a day (08:00, 10:00, 12:00, 14:00, 16:00, 18:00 h). After 30 days, no effect on growth. Result indicates higher importance of in-excess feeding at each meal than number of meals per day 26.
• LAB: FINGERLINGS in round 5 L plastic tanks (40 cm diameter) at density 5 IND/L and either fed continuously by belt feeders or at various times during day or night by hand. After 25 days, highest weight (10.1 g) when fed continuously either a) 51% during 18:00-02:00, 31% during 02:00-10:00, and 18% during 10:00-18:00 (treatment E) or b) 60% during 19:00-23:00 and 40% during 03:00-07:00 (treatment B). Second-highest weight (8 g) when fed continuously during 24 h. Lowest weights (7.4 or 7.2 g) when fed by hand during day only every 4 h either a) four times (46% at 08:00 and then three times 18% each) or b) three times ca 33% (treatment D). Lowest FOOD CONVERSION RATIO under treatment E (0.8 versus 0.9-1.1), lowest feed wastage under treatment B (2.9 versus 6.4-16.1% of total feed applied). Highest feed wastage under treatment D (16.1% of total feed applied). The results indicate an advantage in feeding continuously and in feeding the majority of feed during night 60.

• LAB: JUVENILES in 200 x 50 x 50 cm polyethylene tanks and either fed spatially localised or spatially dispersed once per day. After eight weeks, higher weight under dispersed than localised feeding (192.7 versus 169.7 g). No difference in coefficient of variance (24.4-28.5), but more homogenous size distribution under dispersed feeding. Also, higher specific growth rate (1.5 versus 1.3%/d) and relative weight gain (133.8 versus 106.5%) and lower FOOD CONVERSION RATIO (1.9 versus 2.3) under dispersed feeding. Results indicate advantage of dispersed food distribution probably because dominant individuals cannot monopolise food 61.

• LAB: FRY in 30 L aquaria were either fed ad libitum six times per day, fed once a day at 10% body weight or 5% body weight or not fed at all. At second observation on day 6, higher cannibalism rate the lower the feed supply (ca 6% when starving versus 5.3% at 5% body weight versus 4.3% at 10% body weight versus 0.1% at ad libitum feeding). Difference between conditions decreased with increasing time to ca 3.7% at 5% body weight, 2.5% at 10% body weight, and 0.5% at ad libitum feeding at end of observation period on day 26. Results indicate that feed supply is strongest factor to influence cannibalism (for shelter → D10, for density → D2) 14.

• Feeding and temperature:
  • WILD: of caught individuals, 46% with empty stomach, majority of which in winter: Great Fish and Sundays rivers, Glen Melville dam, South Africa (introduced) 4.

• Daily rhythm:
  • WILD: direct observations and high frequency of full stomachs in individuals caught at dawn in depth <2 m indicate nocturnal feeding. In deeper habitats, feeding was observed day and night: lake Sibaya, South Africa 9.
  • WILD: variation in daily rhythm between individuals (59.5-70 cm, 2.3-3.7 kg) and between seasons: in November, one individual predominantly active at dawn, other at midday, third at dawn and night; in January (warm season), comparatively more active at dusk, night, and dawn. Difference to other studies reporting predominantly night-time activity may be due to better ability of avoiding nets during the day which the current study circumvented using radio telemetry: lake Ngezi, Zimbabwe 7.
  • WILD: highest catches at night, especially within four hours after sunset: lake Victoria 62↶17, Orange river system 63, Olifants river system 64↶17.
  • WILD, JUVENILES-ADULTS: 38.6% of individuals caught during the day had nearly empty stomachs versus 6.1% caught during the night: lake Awassa, Ethiopia 8.
  • WILD, JUVENILES-ADULTS: majority of food items – except for plant material – found in stomachs during night in both wet and dry season: Agbokim waterfalls, Cross River State, Nigeria 12.
  • LAB: FRY in 250 m³ aquarium with Myriophyllum spicatum and sand or in enamel tray. At 2 days, explored aquarium at night, stayed close to vegetation during day. At 6 days, fed throughout 24 h, interspersed with resting 11.
  • LAB: JUVENILES in aquaria (90 x 45 x 45 cm, water depth 30 cm) and either 24 h, 18 h, 12 h, or 6 h dark and light intensity of 150 lux. During six weeks, no difference in time of feeding (via self-feeder): start around 14:00 or 15:00 and decrease around 20:00 h 18.
  • LAB: JUVENILES in 250 L holding tanks under 12 h PHOTOPERIOD often rested on the bottom, only seldomly interspersed with short swimming bouts of 1-2 min. Four JUVENILES were transferred from holding to 235 L experimental tanks (150 x 65 x 30 cm) and fed 3, 24, 48, 72, or 96 h later at either 200-250 lux or in the dark. Fed either pellets (5 mm diameter, 5 mm length, light-grey brown or stained green, red, or blue) or beef liver bits (5 x 5 x 5 mm). Hardly any feeding at 3 h after transfer (→ D13), no difference in feeding pellets or liver in light or dark after 24, 48, 72, or 96 h 19.
• Nocturnal activity: → Daily rhythm.
• Phototaxis: photonegative:
  • LAB: LARVAE in 100 L fibreglass tanks at 24 h light of 100 lux.
    Trial 1: Light beam of 200 lux directed at specific areas at bottom.
    Trial 2: Light of 80-1,400 lux directed at specific areas at bottom.
    Trial 3: Held over night at 0 lux with the help of a black PVC cover. Light of 15-50 lux directed at specific areas at bottom.
    Trials 1-2: Faster rate of dispersal (= at least 50% of individuals dispersed) the brighter the light beam with ca 210-240 s at 80 lux versus ca 20 s at 1,400 lux.
    Trial 3: Faster rate of dispersal the brighter the light beam with ca 50-80 s at 15 lux versus ca 30 s at 30 and 50 lux. Results indicate strong negative phototactic response 17.
• For daily rhythm and...
  ...home range → D15,
  ...depth range → D16.

• LAB: JUVENILES in 30 L glass tanks (30 x 30 x 45 cm) accustomed to 100 lux during 12 h/d were changed to either 24 h light or 0 h light (by using black PVC plastic). After seven days, mainly active swimming (ca 84%) under 0 h light, mainly resting (ca 50%) followed by active swimming (ca 40%) under 24 h light. Over observation period, decrease in resting, increase in active swimming under 24 h light, no change under 0 h light. Initially higher territoriality under 0 h light, but decreased to negligible level by end of observation period. Under 24 h light, territoriality increased 17.
• LAB: JUVENILES in 120 L aquaria (90 x 45 x 45 cm, water depth 30 cm) at either 24 h light or "12 h light, 12 h dark". During six weeks, more swimming activity and less resting under 24 h light than 12 h light, even more so at 150 compared to 15 lux (221.5-229.2 s versus 160.7-165.4 s swimming activity during 5 min under 150 versus 15 lux for 24 h compared to 145.9-201.5 s under 150 or 15 lux for 12 h; 70.8-78.5 s versus 34.6-139.3 s resting during 5 min under 150 versus 15 lux for 24 h compared to 98.5-154.1 s under 150 or 15 lux for 12 h). Also, more aggressive acts like chases, body bites, barbel fights, and lateral displays (35.5 versus 9-11 acts during 5 min under 150 versus 15 lux for 24 h compared to 6-29.5 under 150 or 15 lux for 12 h) and consequently 41.6% more scars or wounds, especially on the side of the body under continuous versus 12 h light 27.
• LAB: JUVENILES in aquaria (90 x 45 x 45 cm, water depth 30 cm) and either 24 h, 18 h, 12 h, or 6 h dark and light intensity – in conditions with light – of 150 lux. After six weeks, no difference in survival. Higher frequency of swimming under 6 h dark compared to 24 h or 18 h (81.9% versus 72.1-72.2%); 12 h in between (80.4%). Increasing number of scars on body (measure for aggression) with increasing PHOTOPERIOD (1.4 at 24 h versus 2.2 at 6 h). Higher glucose under 6 h and 12 h dark than 18 h dark (3.6-3.9 versus 3.1 mmol/L), no difference to 24 h dark (3.6 mmol/L). Higher plasma lactate under 6 h dark (3.7 versus 3.1-3.2 mmol/L), probably due to higher swimming activity and aggression. Higher cortisol under 6 h dark than 24 h or 12 h (158.9 versus 119.6-133.5 ng/mL); 18 h in between (138.3 ng/mL) 18.
• LAB: JUVENILES under either 0 h, 12 h, or 24 h darkness and light intensity – in conditions with light – of 150 lux. After six weeks, no mortality under 24 h dark, increasing with increasing hours of light (6% under 12 h dark, 18% under 0 h dark). Mortality mainly due to cannibalism type II (where the head or parts of the body remain). Lower number of individuals bitten (78% versus 89.8% under 12 h dark, 97.5% under 0 h dark) and lower number of scars per individual (1.1 versus 2.0 under 12 h dark, 3.0 under 0 h dark) under 24 h darkness, increasing with increasing hours of light 25.

• LAB: FRY in 30 L white plastic bins at either 0, 6, 12, 18, or 24 h light (250 lux) a day with the help of a black PVC cover. After 12 days, higher growth rate the less light with highest growth at 0 h light (0.9 mm/d versus 0.73-0.87 mm/d). One of highest condition factors (1.2 versus 0.94-1.15) and tendency of lower mortality (31% versus 34-39%) under 0 h light 17.
• LAB: JUVENILES in aquaria (90 x 45 x 45 cm, water depth 30 cm) and either 24 h, 18 h, 12 h, or 6 h dark and light intensity – in conditions with light – of 150 lux. After six weeks, higher frequency of feeding under 24 h dark compared to the other conditions (6.7% versus 2.1-2.6%). Higher weight under 24 h and 18 h dark (223.9-237.1 g versus 178.5-192.2 g). No difference in specific growth rate (2.4-2.9%/d) and FOOD CONVERSION RATIO (0.7-0.9) 18.
• LAB: JUVENILES under either 0 h, 12 h, or 24 h darkness and light intensity – in conditions with light – of 150 lux. After six weeks, higher weight (92.2 g versus 69.8 g under 12 h dark, 59.5 g under 0 h dark) and higher growth rate (2.0 g/d versus 1.4 g/d under 12 h dark, 1.2 g/d under 0 h dark) under 24 h dark, decreasing with increasing hours of light. No difference in specific growth rate (4.3-5.3%/d) and FOOD CONVERSION RATIO (0.9-1.3) 25.
• For PHOTOPERIOD and growth and feeding frequency → D17.

• Light intensity preference:
  • LAB: FRY in 60 L glass tank divided in compartments between which individuals could move. Acclimated to 100 lux which was changed to either 0, 12, 90, 140, or 260 lux. Every second day, light gradient was reversed. After six days, highest proportion of FRY in compartment with 0 lux (ca 38% versus 15-18%). FRY in compartment with 0 lux larger in size than FRY in other compartments (ca 28 mm versus 23-24 mm) and displayed territoriality in corners or crevices behind air stones, heater tubes, and walls by chasing away conspecifics 17.

• LAB: JUVENILES in 120 L aquaria (90 x 45 x 45 cm, water depth 30 cm) at either 15 or 150 lux. During six weeks, more swimming activity and less resting under 150 versus 15 lux, even more so under continuous versus 12 h light (221.5-229.2 s versus 160.7-165.4 s swimming activity during 5 min under 150 versus 15 lux for 24 h compared to 145.9-201.5 s under 150 or 15 lux for 12 h; 70.8-78.5 s versus 34.6-139.3 s resting during 5 min under 150 versus 15 lux for 24 h compared to 98.5-154.1 s under 150 or 15 lux for 12 h). Also, more aggressive acts like chases, body bites, barbel fights, and lateral displays (35.5 versus 9-11 acts during 5 min under 150 versus 15 lux for 24 h compared to 6-29.5 under 150 or 15 lux for 12 h) and consequently 2.5 times more scars or wounds, especially on the side of the body, under 150 versus 15 lux 27.

• LAB: FINGERLINGS in 3 L glass basins (20 cm diameter, 7.1 cm water depth) and 0.002, 0.02, 0.2, and 1.4 µmoles/m²/s light intensity under white, red, blue, green, and yellow light for 30 min. Under 0.002 µmoles/m²/s, higher number of biting under green light compared to other lightwaves (ca 2.5 versus 0.3-1 bites/IND/min). Under 0.02 µmoles/m²/s, higher number of biting under red light (ca 1.7 versus 0.5 bites/IND/min). Under 0.2 µmoles/m²/s, higher number of biting under green light (ca 1.4 versus 0.2-1 bites/IND/min). Under 1.4 µmoles/m²/s, higher biting under yellow light (ca 1.8 versus 0.2-0.9 bites/IND/min). Only wavelength with low number of bites/IND/min across all light intensities was blue light 35.

• Standard temperature range: 10-41 °C:
  • Observations WILD: 18-28 °C over 20 m, 16-36 °C (rarely 13-41 °C) at <10 cm depth: lake Sibaya, South Africa 9, 10-25 °C: P.K. le Roux dam, South Africa 43, 23-29 °C: lake Chamo, Ethiopia 13, median 16.5-21.4 °C: Guaraguacu river, Parana, Brazil (introduced) 3, 10.8-25.3: Great Fish and Sundays rivers, South Africa (introduced), 14-27 °C: Glen Melville dam, South Africa (introduced) 4, 20.9-21.7 °C: Shesher and Welala wetlands, lake Tana, Ethiopia 48, 10.5-24.5 °C: Darling dam, South Africa (introduced) 2.
• Temperature preference: no data found yet.
• Migration temperature: no data found yet.
• For temperature and...
  ...daily rhythm → D18,
  ...spawning → D1.

• Lower and upper lethal limits:
  • LAB: FRY in one of three 20 L compartments of 75 L aquaria at 10 IND/L and acclimated to 25 °C.
    Experiment 1: 5 day-old FRY were exposed to sudden drop from 25 °C to 15 °C over 6 h or to 5 °C over 20 h. After one day, 0% survival in 5 °C-group, 59.5% in 15 °C- group, decreasing to 1.8% on day 3.
    Experiment 2: 5 day-old FRY were exposed to gradual decrease from 25 °C to 15 °C over two days or to 10 °C over three days. No difference in survival on days 1 and 2 compared to control, but decrease from day 3 on (15 °C: 74.4%, 10 °C: 87.3%, control 25 °C: 95.4%) to almost 0% survival on day 5 (15 °C: 1.7%, 10 °C: 0.2%).
    Experiment 3: see Experiment 2 with 21 day-old FRY. No difference in survival until day 5. Decrease under 10 °C from day 6 on (10 °C: 93.5%, 15 and 25 °C: 97.3-98.5%), 0% survival on day 11.
    Experiment 4: 5 day-old FRY were exposed to cyclic temperature changes of 10 °C (25-15 °C) or 15 °C (25-10 °C). No difference in survival until day 2. Drop of survival on day 3 under 15 °C fluctuations (10 °C: 14.7%, 15 and 25 °C: 98.7-98.8%), no further decrease by day 7.
    Experiment 5: see Experiment 4 with 21 day-old FRY. No difference in survival until day 2. Decrease under 10 °C compared to control beginning on day 3 (10 °C: 97.3%, 15 and 25 °C: 98.5-100%), decrease under 15 °C beginning on day 5 (15 °C: 97.3%, 25 °C: 99.2%). Larger decrease by end of observation period on day 7 under 10 °C than under 15 °C (10 °C: 94.2%, 15 °C: 96.3%, 25 °C: 99.0%) 65.
• Temperature change and stress:
  • LAB: JUVENILES acclimated to 29 °C were transferred to 23 °C, 35 °C, or 41 °C water. After eight weeks, no difference in red blood cell count (1.3-2.3 x 10⁶/mm³), mean corpuscular haemoglobin concentration (18.2-30), mean corpuscular haemoglobin volume (168-202.2), and glucose (74-98 mg%). Lower haemoglobin under 23 °C (4.5 mg%) and 41 °C (2.3 mg%) compared to control (=29 °C: 5.4 mg%). Lower total plasma protein under 23 °C (3.8 mg%) and 41 °C (4.9 mg%) compared to control (6.1 mg%). Results indicate high adaptive ability 66.

• Observations WILD: usually >60% saturation: lake Sibaya, South Africa 9, 65-82% saturation: labe Sibaya, South Africa 11, 6.9 mg/L (at 25 °C): lake Chamo, Ethiopia 13, 5.1-6.2 mg/L: Shesher and Welala wetlands, lake Tana, Ethiopia 48.

• Salinity tolerance:
  • Natural and introduced distribution in fresh water D19 D20.
• Standard salinity range:
  • Observations WILD: 1.0-1.2 mg/L: lake Chamo, Ethiopia 13.

• Standard pH range:
  • Observations WILD: >9: lake Chamo, Ethiopia 13, 7.1-9.2: Great Fish and Sundays rivers, South Africa (introduced), 8-9: Glen Melville dam, South Africa (introduced) 4, 7: Shesher and Welala wetlands, lake Tana, Ethiopia 48.
• pH preference: no data found yet.

• Standard turbidity range:
  • Observations WILD: 87.8-242.0 ntu in the mainstream: Great Fish river, South Africa (introduced), 18.9-26.9 ntu in the mainstem: Sundays river, South Africa (introduced), 129-257 ntu: Glen Melville dam, South Africa (introduced) 4.
  • Observations total dissolved solids WILD: 533.3-890 ppm in the mainstream: Great Fish river, South Africa (introduced), 1,141.8-1,256 ppm in the mainstem: Sundays river, South Africa (introduced) 4.
  • Observations eutrophic WILD: lake Chamo, Ethiopia 38.
• Secchi depth (water transparency): 0.05-3.5 m:
  • Observations WILD: 3-3.5 m: lake Sibaya, South Africa 40, 10-70 cm: P.K. le Roux dam, South Africa 43, 4.6-6.2 cm (0.05-0.06 m): Shesher and Welala wetlands, lake Tana, Ethiopia 48, 16 cm: Darling dam, South Africa (introduced) 2.

• Swimming type: anguilliform:
  • Observations LAB: 20.
• Ontogenesis of swimming behaviour:
  • LAB: 0-48 hph (hours post hatching) LARVAE and small FRY (until 168 dph) in 75 x 25 mm aquarium. Until 8 hph and 5.5 mm TOTAL LENGTH, non-swimming movements, for example circling around yolk sac. At 10 mm TOTAL LENGTH (after yolk sac absorption), reduced amplitude compared to 5.5 mm and swimming movements resembling those of ADULTS 20.
• For swimming and stocking density → D2.

• Absolute swimming speed:
  • Observations WILD: range from 478 m/h during 2-3 h to 105 m/h for 6 h: lake Ngezi, Zimbabwe 7.
• Relative swimming speed:
  • Observations LAB, FRY: at size 8-21 mm swam at speeds of 1.5-9.5 TOTAL LENGTH/s 20.
• Swimming speed and temperature: no data found yet.

• Standard velocity range: no data found yet.
• Velocity preference:
  • LAB: 15 day-old FRY in circular 4 L tanks of either 1.9 (0.04 m² x 0.1 m depth; "deep" tank), 14.5 (0.1 m² x 0.03 m depth; "shallow" tank), or 150 (0.6 m² x 0.006 m depth; "shallow tank") diameter-to-depth-ratio at 25 IND/L (2,857 IND/m², 1,613 IND/m², 719 IND/m², or 157 IND/m² respectively) and inflow pipe at flow rates of 0.6-4.6 L/min. For comparison: 216 L tank with diameter-to-depth-ratio 10. Flow rates resulted in faster velocities in deep than shallow 4 L tanks. Velocity at periphery faster than in center. In tanks with 1.9 and 10 diameter-to-depth ratio, central and peripheral current velocities increased with flow rate, whereas in shallow tanks, central velocity was constant, did not reach or go beyond 2 cm/s. This calm center may provide a refuge. Linear relationship between fish size and maximum current velocity which allows individual to still maintain station without swimming (e.g., ca 13 mm fish and ca 0.7 cm/s, ca 96 mm fish and ca 9.5 cm/s) 24.
• Velocity and temperature: no data found yet.

• WILD, individuals (59.5-70 cm, 2.3-3.7 kg): local (<40 m), moderate (40-200 m) and long-term movements (>200 m). At least one long-term movement (>200 m) per day, increasing in July-August and January. Individual difference in number of long-term movements: most active individuals with up to 13 and 20 long-term movements during 24-36 h observation periods. Range from one individual with 1.1 km movement within 2-3 h to 5.7 km within 6 h. Stayed in restricted area (1,600-2,000 m²) after feed detection for days to months: lake Ngezi, Zimbabwe 7.
• WILD, individuals (38-100 cm, 520-6,320 g): range 2-70 ha, wider during midmorning and after midnight: Glen Melville dam, South Africa (introduced) 4.

• Depth range in the wild:
  • Observations WILD: FRY and JUVENILES were caught in marginal and littoral zones: lake Sibaya, South Africa 40, majority of recaptured JUVENILES or non-spawning ADULTS in littoral: lake Edward on border between Democratic Republic of the Congo and Uganda 39, individuals (59.5-70 cm, 2.3-3.7 kg) moved in depths of 0.5-10 m: lake Ngezi, Zimbabwe 7, Shesher wetland 1.8 m max, Welala wetland 2.5 m max: lake Tana, Ethiopia 48.
  • For depth and spawning → D1.
• Depth in cages or tanks: no data found yet.
• Depth preference: <3-10 m:
  • Observations WILD, individuals (38-100 cm, 520-6,320 g): preferred river mouth (<3 m) and upper section (3-10 m) of Glen Melville dam, South Africa (introduced) 4.
• Depth and daily rhythm:
  • WILD: JUVENILES-ADULTS entered shallower water (<2 m) predominantly at night to feed; frequency of catches in deeper waters approximately equal during day and night: lake Sibaya, South Africa 9.
  • WILD, individuals (38-100 cm, 520-6,320 g): in river mouth (<3 m) after midnight, in upper section (3-10 m) during midmorning: Glen Melville dam, South Africa (introduced) 4.
• Depth and low temperatures: no data found yet.
• Depth and high temperatures: no data found yet.
• Position in habitat and age:
  • WILD: LARVAE and small JUVENILES (<50 mm) in rootstock of semi-aquatic plants in shallow water <30 cm, JUVENILES of 50-200 mm occasionally visit open water at night, JUVENILES of 200-400 mm in marginal pools 50-150 cm, ADULTS (>350 mm) in deeper habitats: ADULTS of mode 509 mm in terrace habitats 0.8-3.8 m, ADULTS of mode 499 mm in gradual sloping habitats and deep slopes until 15 m, ADULTS of mode 603 mm in profundal zone until maximum depth of lake of 40 m: lake Sibaya, South Africa 9.
  • WILD: JUVENILES of 10-20 cm TOTAL LENGTH in 25 cm shallow water, JUVENILES-ADULTS in lagoons of maximum 4 m or riverine area: Elephant marsh of river Shire, Malawi 10.
• Depth and light intensity: no data found yet.
• Depth and noise: no data found yet.
• Depth and threat: no data found yet.

• JUVENILES and ADULTS migrate within fresh water, ADULTS migrate to spawning grounds:
  • WILD, JUVENILES: of 40 returns within 300 days, 10 JUVENILES had moved upriver, five downriver, 25 were retrieved close to release site: Elephant marsh of river Shire, Malawi 10.
  • WILD: individuals (59.5-70 cm, 2.3-3.7 kg) performed long-distance movements especially in July-August (cold season) and January (warm season) within lake and upriver: lake Ngezi, Zimbabwe 7.
  • WILD: JUVENILES-ADULTS migrated to waterfalls April-July (presumably to spawn) and October-December: Agbokim waterfalls, Cross River State, Nigeria 12.
  • WILD: majority of individuals caught in Ribb river mouth was JUVENILES (87.5%), majority caught in Shesher and Welala wetlands close to lake Tana, Ethiopia were ripe or spent (32% ripe, 22.5% spent) indicating migration of JUVENILES (46.5%) to wetlands to feed and of ADULTS to spawn 48.
  • WILD: ADULTS moved to lake shore or inundated grasslands to spawn: lake Sibaya, South Africa 11.
  • WILD: ADULTS probably migrated from littoral to wetlands, river channels, and river mouths to spawn: lake Edward on border between Democratic Republic of the Congo and Uganda 39.

• Observations time from fertilisation until hatching WILD: 24-25 h: lake Sibaya, South Africa 11.
• Observations time from fertilisation until hatching LAB: most at 24-25 h, some at 36-48 h (at 19-33 °C) 11, 36 h (28 °C) 14, 24-28 h 17, 24 h (at 30 °C) 23, ca 24 h (26 °C) 67, 24-26 h (27 °C) 31, 24-48 h (26 °C) 68.
• Observations size WILD: 1.6-1.9 mm: lake Sibaya, South Africa 11.
• Observations size LAB: 105-125 µm 68.
• Observations weight: no data found yet.

• Observations age and TOTAL LENGTH at yolk sac absorption LAB: 3 days 31, complete by 60 hph (hours post hatching) and 8.5 mm (previous observation time: 48 hph) 20, 3 days 69.
• Observations TOTAL LENGTH and weight LAB: 3.6 mm at hatching 11, 4.4 mm after hatching, 5.2 mm and 1.5 mg 68, 4 mm at hatching 20.

• Observations age, TOTAL LENGTH, and weight at beginning of exogenous feeding LAB: 60 h after hatching, 6.2 mm 11, 4 days after hatching 14, 48-72 h after hatching 17, 48 h after hatching 23, 48 h after hatching 60, 7.3 mm and 2.8 mg 58, 48 h after hatching 20.
• Observations age and TOTAL LENGTH WILD: 3 weeks, mode 18 mm: lake Sibaya, South Africa 40.
• Observations age, TOTAL LENGTH, and weight LAB: body looked like that of ADULTS at 14 days and 12.1 mm 11, 3 weeks and 18.2 mm, 4 weeks and 22.2 mm, 70 days and 41 mm 11, 5 weeks, 27 mm, and 0.2 g 17, beginning of air breathing at 12-14 days after beginnng of exogenous feeding, i.e. at days 14-16 23, 8-21 mm 20, 10 days, 0.003 g 69, 6 weeks, 42.9 mm, 0.7 g 22, 3 days, 5.9 mg 70.

• Fingerlings: FRY with working fins, the size of a finger, 25-75 days, 4-8 cm, 1 g:
  • Observations age, TOTAL LENGTH, and weight LAB: 25 days, 1 g 60, 2.5 months, 5-8 cm 16, 4-5 cm 35.
• Juveniles: fully developed to beginning of maturity, 0-2 years, 8-60 cm, 7.5-750 g:
  • Observations age, TOTAL LENGTH, and weight WILD: 1-2 years, 18-30 cm: Elephant marsh of river Shire, Malawi 10, 1 year, mode 240-260 mm, range 80-410 mm, mean 97 g: lake Sibaya, South Africa 40, 16.3-35 cm: lake Awassa, Ethiopia 8, 17.1-44.7 cm: lake Chamo, Ethiopia 38, 200-600 mm, 150-750 g: Vembanad lake, Kerala, India (introduced) 45.
  • Observations age and weight FARM: 5 months, females 117 g, males 128 g 71, 27.5 g 50.
  • Observations age, TOTAL LENGTH, and weight LAB: 59.6 mm 14, 137 days, 100 g 72, 12-18 cm 21, 286 mm and 186.7 g 15, 47 weeks and 55 g 27, 82 g 61, 69.2 g 18, 4 months, 20-24 cm standard length, 80-95 g 19, 58.9 g, 75.5 g 73, 7.5 g 33, 192.8 g 29, 9.9 g 25, 13.2 cm 74.
• Sexual maturity for 50% of JUVENILES: 0-6 years, 16.5-92 cm for males, 14.6-74+ cm for females:
  • Observations age and TOTAL LENGTH WILD: 50% of males mature at ca 2-3 years, 32 cm, 50% of females at 29 cm, smallest mature female: 21 cm: Elephant marsh of river Shire, Malawi 10, some at the end of first year, 200-250 mm, median size of 50% population 330 mm for females, 340 mm for males, majority of population mature at end of second year: lake Sibaya, South Africa 11, 50% maturity at ca 4-6 years, for males at 820-920 mm, for females at >740 mm: P.K. le Roux dam, South Africa 43, smallest ripe female 34 cm, smallest ripe male 33 cm: lake Awassa, Ethiopia 8, males 16.5 cm, females 14.6 cm: Agbokim waterfalls, Cross River State, Nigeria 12.
  • Observations age LAB: matured at 10 months 68.

• Observations age, TOTAL LENGTH, and weight WILD: >350 mm: lake Sibaya, South Africa 9, 3 years, 38 cm, largest specimen 100 cm and 8.5 kg: Elephant marsh of river Shire, Malawi 10, 7 years: mean 648 mm and 1,612 g for females, mean 695 mm and 1,873 g for males, 8 years (males): mean 726 mm, mean 2,107 g, largest females: 951-1,036 mm and 5.7-7.8 kg, largest males: 916-1,088 mm and 5.5-8.8 kg: lake Sibaya, South Africa 40, 8 years, 800-1,080 mm: P.K. le Roux dam, South Africa 43, 1-2 kg: Hula nature reserve, Israel 44, 38-110 cm, 52.5 cm and 1,100 g: lake Awassa, Ethiopia 8, female 435 mm and 686 g, female 515 mm and 1,044 g, female 700 mm and 2,562 g: Guaraguacu river, Parana, Brazil (introduced) 3, female 484 mm and 797.6 g, female 650 mm and 1,780.0 g, female 850 mm and 3,938.8 g: lagoa Encantada, Bahia, Brazil (introduced) 47, 55.2-109 cm: lake Chamo, Ethiopia 38, 650-750 mm, 1,400-2,500 g: Vembanad lake, Kerala, India (introduced) 45, largest female 13 years, 1,240 mm, oldest female 21 years and 885 mm, oldest males 25 years and 840-1,074 mm: Darling dam, South Africa (introduced) 2, 8 years, range 493-554 mm: Okavango delta, Botswana 42.
• Observations age and weight FARM: 9 months, 334 g 75, 450-500 g 31.
• Observations weight LAB: 3.3-3.5 kg 76.

• Observations WILD: 0.9:1-1.2:1: Elephant marsh of river Shire, Malawi 10, 1.4:1 in general, 1:1.1 until 500 mm TOTAL LENGTH, 3.8:1 from 600 mm TOTAL LENGTH on: lake Sibaya, South Africa 40, 1:1: lake Awassa, Ethiopia 8, 1:0.7: Okavango delta, Botswana 42.
• Observations LAB: 40:60 to 50:50 68.

• Growth: annulus formation once a year in May-October:
  • Observations annulus formation WILD: on vertebrae during May-August (cold season): Elephant marsh of river Shire, Malawi 10, on pectoral spines in October: P.K. le Roux dam, South Africa 43, on otoliths once a year: Darling dam, South Africa (introduced) 2, on otoliths in July-September: Okavango delta, Botswana 42.
• Natural growth rate:
  • WILD, LARVAE-JUVENILES: approximate 18-38 mm (mean 24 mm) growth per month since hatching, amounting to modal size of 144 mm at 6 months. Yearlings' mode size 240-260 mm. Pectoral spine readings: first ring at modal 400-410 mm, probably at end of second year: lake Sibaya, South Africa 40.
  • WILD, LARVAE-ADULTS: ca 11 cm in first 4 months, at 15-16 cm at 6 months, ca 20 cm at 1 year, ca 30 cm at 2 years, ca 38 cm at 3 years, ca 42 cm at 4 years (ca 40 cm for females), ca 50 cm at 5 years (ca 45 cm for females), ca 60 cm at 6 years (ca 50 cm for females), ca 70 cm at 7 years: Elephant marsh of river Shire, Malawi 10.
  • WILD, LARVAE-ADULTS: 204-213 mm growth in the first year, then decrease and increase inconsistently along 88-129 mm for females and 92-163 mm for males in consecutive years, in year 8: 49 mm for females, 61 mm for males: P.K. le Roux dam, South Africa 43.
  • WILD, LARVAE-ADULTS: fast growth (up to 300+ mm) in first year, then large variation probably due to available food, e.g. biggest individual of 0-8 year olds at 4 years (731 mm), smallest at 2 years (352 mm): Okavango delta, Botswana 42.
  • WILD/FARM: JUVENILES from six strains caught all over Nigeria and placed in 1 m³ hapas at 25 each and fed 3% body weight twice daily. After 10 weeks, weight gain 372.7-635.1%, 1.6-1.9 FOOD CONVERSION RATIO, 1.0-1.2% daily specific growth rate, 76-96% survival rate. Domesticated strain had higher weight gain (647.7%) and lower FOOD CONVERSION RATIO (1.4), similar specific growth rate (1.2%/d) and survival (70%), probably due to selective breeding. Results support previous findings of considerable growth variation in C. gariepinus within and between strains 41.

• Observations bimodal pattern WILD: largest females: 95.1-103.6 cm and 5.7-7.8 kg, largest males: 91.6-108.8 cm and 5.5-8.8 kg: lake Sibaya, South Africa 40.
• Beginning of noticeable size difference:
  • WILD, ADULTS: no difference noted between sexes in first 3 years, thereafter slightly lower growth rate in females (→ D21) probably due to increased energy expenditure for growth of oocytes: Elephant marsh of river Shire, Malawi 10.
  • WILD, ADULTS: growth approximately similar for males and females in first 3 years, then males grew faster: lake Sibaya, South Africa 40.
  • LAB: JUVENILES in 140 L aquarium. After 100 days, faster growth in males compared to females (322 g versus 302 g) 72.
  • LAB: at 12 months, males 140.5 g, females 123.5 g 68.

• LAB: FRY in 120 L tanks (90 x 45 x 45 cm) at density of 36 IND/tank and sorted into four groups: homogeneous low weight (L, mean 83 g, range 66.5-94.9 g), homogeneous medium weight (M, 140 g, 125.4-154.9 g), homogeneous heavy weight (H, 198 g, 184.5-214.1 g), heterogeneous group with one third of other groups (HET, 139.7 g, 65.2-214.3 g). After 27 days, no difference in growth rate (25.6-28.1 g/kg^0.8/d), feed intake (22.7-24.4 g/kg^0.8/d), and FOOD CONVERSION RATIO (0.8-0.9 (g/kg^0.8/d)/(g/kg^0.8/d)). Among similar-sized FRY in L, M, H, and HET groups, no difference in growth rates. Results indicate no compensatory growth of low weight FRY and no large effect of social hierarchy on growth in the sense that larger (supposedly dominant) individuals suppress growth in small (supposedly subordinate) conspecifics 30.

• Growth in polyculture:
  • FARM: JUVENILES (5 g) in 150 m² (1.5 m deep) earthen ponds stocked at 1:3 ratio with Oreochromis niloticus at low (30,000 IND/ha), medium (60,000 IND/ha), or high density (90,000 IND/ha). Deliberately stocked 30 days after O. niloticus (5 g mean weight) so that these were too big for C. gariepinus to prey on. After 240 days, higher growth of O. niloticus under poly- than monoculture (75,460 g weight gain versus 66,879 g), specifically under low stocking density (204.7 g versus 141.6 g under high, 132 g under medium stocking density). Also, higher survival (98.1% versus 86.6-89.3%) and higher specific growth rate (1.6% versus 1.4%) but higher FOOD CONVERSION RATIO (3.5 versus 2.6-3.1) under low density. In C. gariepinus, higher weight gain under low than medium and high stocking density (579.5 g versus 519.2 under medium, 341.3 under high density). Also, higher survival (97.9% versus 89.5-96.6%) and higher specific growth rate (2.2% versus 2% in high density) but higher FOOD CONVERSION RATIO (3.5 versus 2.6 under high density). Results indicate that polyculture with C. gariepinus is beneficial for O. niloticus probably because C. gariepinus preys on O. niloticus juveniles that would otherwise compete for food with adults 77.
  • LAB: JUVENILES (8.5 g) in 40 L circular flow-through tanks either reared in monoculture or duo- or trioculture with other Clariid Heterobranchus longifilis and hybrid (male Heterobranchus longifilis x female C. gariepinus). After eight weeks, best weight gain, specific growth rate, and FOOD CONVERSION RATIO in monoculture than duoculture or trioculture (monoculture: weight gain 1,593.3%, specific growth rate 5.1, FOOD CONVERSION RATIO 1.8; duoculture with H. longifilis: 1,074.7%, 4.4, 1.8; duoculture with hybrid: 1,195.5%, 4.5, 1.8; trioculture: 791.9%, 4, 2.8). Best growth under monoculture of hybrid with weight gain 1,980.3%, specific growth rate 5.4, FOOD CONVERSION RATIO 1.8 78.
• Growth and tank design:
  • LAB: 15 day-old FRY:
    Experiment 1: in circular 4 L tanks of either 0.03 m or 0.1 m depth at 6.3 IND/L (1,101 IND/m²). After 20 days, no difference in wet weight.
    Experiment 2: in circular tanks of 0.1 m depth of either 4 or 16 L at 25 IND/L (2,857 or 2,878 IND/m² respectively). After 20 days, no difference in wet weight.
    Experiment 3: in circular 4 L tanks of either 1.9 (0.04 m² x 0.1 m depth), 4.3 (0.06 m² x 0.07 m depth), 14.5 (0.1 m² x 0.03 m depth), or 150 (0.6 m² x 0.006 m depth) diameter-to-depth-ratio at 25 IND/L (2,857 IND/m², 1,613 IND/m², 719 IND/m², or 157 IND/m² respectively). After 20 days, better growth in shallower than deeper tanks (ca 800 mg in 14.5 versus ca 580 mg in 1.9 diameter-to-depth ratio). No difference in mortality due to cannibalism (0-5.5%) or territoriality (0-7.0%) in experiments 1-3.
    Results indicate advantage of shallower tanks which also reduce the distance to travel for – and therefore the energetic costs associated with – air breathing 24.

• FARM: survey among 58 fish farmers in Nigeria revealed 91.4% had noticed abnormalities in the past – either rarely (41.4%), occasionally (32.8), or very often (25.9%). Mostly on the head (69% of farmers) but also on the body (20.7%; spinal deformities, tumours, fin deformities, skin erosion) during all life stages (8.6% FINGERLINGS, 27.6% post-FINGERLINGS, 17.2% JUVENILES, 10.3% post-JUVENILES, 36.2% ADULTS). Resulting in decreased growth (93.1% of farmers) and affecting survival to market (at ca 6 months) in 0-100% of individuals, mostly in 50% (25.9% of farmers) to 75% (34.5% of farmers) of individuals 79.

• LAB: sperm of four males was cryopreserved in liquid nitrogen and either methanol or dimethyl-sulfoxide as cryoprotectant in straws of either 0.3, 0.5, or 1.3 mL volume for 3-12 months, then thawed and mixed with eggs from females that were induced to ovulate by injecting Gonadotropin Releasing Hormone Analogue. No difference in fertilisation (62-82.9% versus 83.6%), hatching (43-53% versus 59.1%), and malformation rate (32.8-44.2% versus 28.3%) compared to control. Mainly deformities of the tail but also head. Mainly diploidy (56 chromosomes) in all groups including control (51-69% versus 58-63% control), but certain proportion of LARVAE from cryopreserved sperm were haploid (28 chromosomes, 1-6%), triploid (up to 84 chromosomes, 8-13%), or tetraploid (112 chromosomes, 1-7%) and certain control LARVAE displaying malformations were triploid (12%) or tetraploid (4%). Results indicate that haploidy is caused by cryopreservation 67.
• For deformity rate and stress → D27.

• Nest building and substrate: no data found yet.
• Nest building and water velocity: no data found yet.
• Nest building and water depth: no data found yet.
• Nest building: no data found yet.
• For spawning → D28.

• Courtship sequence:
  • WILD: pre courtship: ADULTS assembled at dusk in groups, relatively motionless. More active with beginning of darkness: mildly aggressive displays, slapping of caudal fin and head on water surface (which could be heard 300 m away on silent nights), ritualised aerial breathing: releasing air bubles, smacking noises with mouth; severely aggressive butting of urogenital opening, head, or abdomen, nipping, biting, chasing, rarely: mouth-fighting, tearing with the pectoral spine. Probably exclusively performed by males. Males <400 mm rarely took part. Within 60-80 min since beginning, aggression decreased, courtship and mating increased.
    Courtship: males established who is eligible to court by fighting. Winner swam near female and butted softly on body. Receptive female moved to shore and into inundated grassland. Male followed and fought attacking competitors on the way. Butted female again slightly on abdomen, head, and tail. In 50-300 mm deep water, both quivered, butted each other (the female on the urogenital opening of the male, the male on the abdomen of the female): lake Sibaya, South Africa 11.
• Courtship duration:
  • WILD, ADULTS: 1-2 h for intensive courting and mating: lake Sibaya, South Africa 11.

• Spawning substrate: submerged vegetation:
  • Observations WILD, ADULTS: in shallow pools protected by Phragmites mauritianus or Typha latifolia, at the shore among Juncus krausii and Panicum meyerianum, in pools with Sporobolus virginicus, Dactyloctenium geminatum, Andropogon amplectens, Cyperus spp., in flooded grassland with Eragrostris capensis, Cladium spp., Harpochloa spp.. Usually emergent or submerged terrestrial plants, not aquatic plants (e.g., Myriophyllum spicatum, Potamogeton pectinatus, P. schweinfurthii, Ceratophyllum demersum) neither submerged rocks or branches. Eggs on sand or in detritus died, eggs individually attached to leaves or stems survived: lake Sibaya, South Africa 11.
  • WILD, ADULTS: breeders were observed in shallow water among reed stems: Elephant marsh of river Shire, Malawi 10.
• Spawning season: summer:
  • Observations WILD, ADULTS: November-March, peak at end of December (warm season): Elephant marsh of river Shire, Malawi 10, September-February (warm season): lake Sibaya, South Africa 11, May-August: Hula nature reserve, Israel 44, February-June (rainy season): lake Awassa, Ethiopia 8, May-July: Agbokim waterfalls, Cross River State, Nigeria 12.
• Spawning (day)time:
  • Observations WILD, ADULTS: mainly at night, peak at 20:00-02:30 h, usually on second or third day of new moon or last quarter, i.e. at dark nights: lake Sibaya, South Africa 11.
• Spawning temperature:
  • Observations WILD, ADULTS: >18 °C, usually >22 °C: lake Sibaya, South Africa 11, 25-28 °C: Hula nature reserve, Israel 44.
• Spawning salinity: fresh water → D29 D30.
• Spawning and water velocity: no data found yet.
• Spawning depth:
  • Observations WILD, ADULTS: differed with water level: from <300 mm pool to which individuals moved via 60 mm channel over shore or adjacent pools of <400 mm to flooded grassland: lake Sibaya, South Africa 11.
• Spawning density:
  • Observations WILD, ADULTS: increased density of 10-20 IND/125 m shoreline (range 0-30 IND/125 m) after rainfall, 30-60 IND/125 m shoreline (range 20-300+ IND/125 m) shortly before spawning; ca 50-800 ADULTS within one spawning run: lake Sibaya, South Africa 11.

• Spawning sequence:
  • WILD, ADULTS: male positioned his body around female's head in U shape. In 28 of 30 mating pairs, male was larger than female by average 118 mm (range 80-240 mm) which accommodates closer contact of the genital papillae. Shortly afterwards both became still and remained in this quasi amplexus for 17-18 s (range 12-20 s). Male probably released sperm during stiffening and arching, moved body over female's body. Female stiffened and arched, burrowed head underneath male's body into substrate, and released eggs in cloud together with air bubbles. Female anchored in substrate and immediately swished tail to mix sperm and eggs and distribute them. After rest, both partners could resume courting and mating for another 1-4 times until interrupted by competing male. Interval between matings increased from 40 s to >10 min: lake Sibaya, South Africa 11.
• Spawning duration:
  • WILD: 1-2 h for intensive courting and mating: lake Sibaya, South Africa 11.

• LAB: ripe ADULTS in hapas of 1.5 x 1 x 1 m in concrete tanks and either 25, 50, or 75 cm water depth at density of either 2, 4, or 6 pairs. Highest spawning success at 2 pairs and 25 or 50 cm water depth, decreasing with increasing density and increasing water depth. Higher spawning duration in 25 compared to 50 cm depth (4 days versus 1 day) indicate advantage of 25 cm depth 80.

• Number of spawns: no data found yet.
• Fecundity per spawn:
  • Observations absolute fecundity WILD, ADULTS: 2,000-120,000 eggs/female: Elephant marsh of river Shire, Malawi 10, 6,000-163,000 eggs: lake Sibaya, South Africa 11, 8,800-650,000 eggs absolute fecundity: lake Awassa, Ethiopia 8, range 12,576-103,984 eggs, mean 12,151 eggs for females of 23.8 cm standard length and 167.6 g upstream, mean 56,203 eggs for females of 65.9 cm standard length and 878.8 g downstream: Agbokim waterfalls, Cross River State, Nigeria 12.
  • Observations relative fecundity WILD, ADULTS: 600-1,400 g eggs/wet weight female: Elephant marsh of river Shire, Malawi 10, 435-1,176 eggs/g wet weight, mean 669 eggs/g 8, 339-408 eggs/g body weight: Agbokim waterfalls, Cross River State, Nigeria 12.
• For fecundity and importance of olfaction → D31.

• Fecundity and stress:
  • LAB, ADULTS: females were transferred from holding tanks to laboratory 6 h before hormone treatment (with 0.5 mL/kg body weight Ovaprim) and either kept out of the water for 1 h, 2 h, 3 h, or 4 h or not exposed to air but with pre-existing bruises. Females were stripped and the eggs mixed with milt. In females with bruises from handling, lower number of eggs released (24,150 versus 75,558), lower fertilisation rate (49% versus 75%), hatching rate (34.5% versus 65%), survival of the LARVAE (62.5% versus 77.5%), and higher deformity rate (24.9% versus 13.2%) compared to control.
    In females exposed to air, increase in stress effects with increasing time out of the water with one of the lowest number of eggs released (15,680), fertilisation rates (47.5%), hatching rates (29.5%), survival rates of the LARVAE (52.5%), and one of the highest deformity rates (28.1%) under 4 h air exposure compared to control.
    Another group of females were transported for 2-3 h in 25 L containers without oxygen to laboratory before hormone treatment. Lower number of eggs released (28,185 versus 75,558), lower fertilisation rate (52.5% versus 75%), hatching rate (34.5% versus 65%), survival of the LARVAE (67.5% versus 77.5%), and higher deformity rate (25.5% versus 13.2%) compared to control 31.
  • LAB, ADULTS: females in 200 L indoor plastic tanks at 3 IND/tank and exposed to either one of stresses: starvation for 10 days, poor handling causing bruises, one hour out of water, five hours out of water. Then injected with GnRHa and dopamin antagonist Domperidone. Three more experimental groups after injection: chasing with net for 2 min each hour, two-fold increase in density, three-fold increase in density. Higher latency until ovulation compared to control (11.5 h) under starvation for 10 days (12.7 h), poor handling (12.2 h), chasing (12.8 h), and two-fold increase in density (13 h). Females were stripped of eggs. No difference in ovulation compared to control (50-100% versus 100%). Lower egg weight per female (47.5-55.8 g versus 77.2 g) and lower number of eggs compared to control (32,730-39,975 eggs versus 54,935 eggs) in all experimental groups. Eggs were mixed with milt. Lower fertilisation rate compared to control (81.1%) under starvation (70%), chasing (71.2%), two-fold (70.1%), and three-fold increase in density (66.5%). Lower hatching rate compared to control (83.7%) under starvation (64.1%), poor handling (60.9%), chasing (71.9%), two-fold (65.1%), and three-fold increase (71.8%). Higher deformity rate compared to control (2.7%) under starvation (14.1%), poor handling (5.6%), chasing (5.1%), two-fold (7.6%), and three-fold increase (7%). Higher number of dead LARVAE compared to control (2.8%) under starvation (7.1%), poor handling (5.2%), one hour out of water (5.9%), three-fold increase (5.1%). Lower survival rate compared to control (94.3%) under starvation (78.8%), poor handling (89.1%), one hour out of water (89.6%), chasing (90.5%), two-fold (88.2%), and three-fold increase (87.9%) 36.

• Fecundity and temperature manipulation: no data found yet.
• Fecundity and hormone treatment:
  • FARM: ADULTS kept in outdoor ponds. Despite yearlong injection with suspended Carp pituitary, relative fecundity fluctuated: was lower during dry colder season (ca 22 °C; 1.3% in August versus 14.3% in January). Indicates mixture of various influences on spawning 81.
  • LAB, ADULTS: males were kept individually in 120 L tanks and injected with hormones:
    Experiment 1: either 8 mg/kg homologous pituitary suspension from C. gariepinus (Clarias-PS), 60 µg/kg mGnRHa, 20 µg/kg mGnRHa plus 5 mg/kg pimozide (mGnRHa-PIM) each dissolved in 0.9% NaCl and applied at 0.5 mL/kg body weight or 0.9% NaCl (control). After 24 h, no hand stripping possible, testes still small possibly due to strain selected for delayed maturity.
    Experiment 2: either 8 mg/kg Clarias-PS, 8 mg/kg pituitary suspension from Carp (Carp-PS), 60 µg/kg mGnRHa each dissolved in 0.9% NaCl and applied at 0.5 mL/kg body weight or 10 µg/kg sGnRHa plus 5 mg/kg domperidone (0.5 mL/kg Ovaprim), 20 µg/kg sGnRHa plus 10 mg/kg domperidone (1.0 mL/kg Ovaprim) or 0.9% NaCl (control). After 24 h, stripping possible under Clarias-PS, 0.5 mL/kg Ovaprim, and 1 mL/kg Ovaprim, but fluid contained no moving sperm cells.
    Experiment 3: 8 mg/kg and 10 mg/kg Carp-PS with 48 h interval (or saline). Stripping was possible 12 h later, not 24 h later. Very low sperm motility. In intratesticular semen, higher volume than in control (27.7 mL versus 7.3 mL) and higher cells/kg (38.9-53.9 x 10⁹ cells/kg versus 15.8 x 10⁹ cells/kg), but lower spermatocrit (20.5% versus 47.0%) and no difference in cells/mL (7.5-10.1 x 10⁹).
    Experiment 4: 8 mg/kg Carp-PS and after 48 h either 5 mg/kg Carp-PS or 10 µg/kg sGnRHa plus 5 mg/kg domperidone (0.5 mL/kg Ovaprim). After 24 h, 0.5 mL/kg Ovaprim (PS-PS-ovp or PS-ovp-ovp). After 24 h, stripping was possible, but very low sperm motility. In intratesticular semen, no difference in volume (17.9-32.5 mL), tendency of lower spermatocrit (8.8-28.9% versus 52.6%) and lower cells/mL (2.2-6.7 mL versus 10.8 mL) compared to control. No difference in hatching rates (66.0-88.5%).
    Results indicate no spontaneous release of semen and almost impossible hand stripping possibly due to anatomical blockage. Two consecutive injections with Carp pituitary suspension increased number of sperm cells/kg but only in intratesticular semen 76.

• WILD, ADULTS: after mating, female wiggled tail fin to disperse eggs and increase probability of survival, as eggs in clumps (determined in LAB) and on sand or in detritus died. Also, spawning in recently inundated grasslands probably decreases threat of predators: lake Sibaya, South Africa 11.

• LAB: FRY in groups of five individuals in 26.5 x 16.5 cm transparent polystyrene aquaria (water depth 20 cm) under natural PHOTOPERIOD (max. 1,472 lux). FRY were presented with a green and red plastic plate (12.4 x 6 cm) and received reward (glued-on feed) if chose correct colour. Had learned colour discrimination when visited correct plate even without reward more often than incorrect one.
  Had learned colour vision (i.e., chromaticity instead of brightness) when visited correct plate more often than two simultaneously presented grey plates. The latter was unsuccessful in three aquaria when paired with grey plates of 40:60 and 60:40 white:black ratio 22.
• LAB: four JUVENILES were transferred from holding to 235 L experimental tanks (150 x 65 x 30 cm) and fed 3, 24, 48, 72, or 96 h later at either 200-250 lux or in the dark. Fed either pellets (5 mm diameter, 5 mm length, light-grey brown or stained green, red, or blue) or beef liver bits (5 x 5 x 5 mm).
  Experiment "1+1": If given a pair with one of either colour, JUVENILES first ate blue when combined with green or red, first ate red when combined with green, first ate liver when combined with blue.
  Experiment "1+3": If given three blue pellets plus one red or three red pellets plus one blue, JUVENILES first ate blue. If given three blue pellets and one liver bit, ate liver first.
  Experiment "5+5": If given five pellets each of two colours, JUVENILES first ate blue when combined with green or red (and the other colour last), first ate red when combined with green (and green last).
  Experiment "1+1+1": If given one pellet of each of three colours, JUVENILES first ate blue, then red, then green pellets.
  Experiment "15+15+15": If given 15 pellets of each of the colours in darkness, no difference in preference (i.e., pellets left) 19.

• Vision and feeding, swimming, "walking" across land:
  • LAB: colour vision – or vision in general – is not needed in often muddy habitats with poor visibility, but feeding and swimming during daytime (not just nocturnally; → D18) and "walking" across land suggest otherwise 22.
  • LAB: FRY in 200 x 90 x 90 mm glass aquaria had eyes cauterised and were presented with 50 large Daphnia. During 30 min, no difference in mean percentage Daphnia consumed compared to control (40.3-41.0%). Result indicates non-importance of vision for foraging 14.
  • LAB, JUVENILES: large cerebellum (25% volume ratio), prominent optic tectum (16% volume ratio), indicating high importance of vision and adaptation to swimming during day and night 74.
  • For (minor) importance of vision and...
    ...feeding and preference in food pellet colour → D6,
    ...and ovarian growth → D31.

• LAB: FRY in 6 L aquarium tanks coated either black, blue, or green and either with shade (broken plastics) or not. After 15 days, higher survival in black tanks compared to other colours (93.1-93.4% versus 78.4-87.5%) 70.

• LAB: FRY in 6 L aquarium tanks coated either black, blue, or green and either with shade (broken plastics) or not. After 15 days, tendency of better growth in tanks with shade than without, tendency of better growth (31 mg versus 27.9-29.5 mg) and specific growth rate (1.6 versus 1.3-1.5%/d) in black tanks with shade 70.

• LAB: FINGERLINGS in 3 L aquaria were fed red agar-agar pellets laced with taste substances (5% citric acid = sour, 10% sodium chloride = salty, 10% calcium chloride = bitter, 10% sucrose = sweet) or amino acids. Higher consumption of taste pellets than control (49.1-81.0% versus 9.5%), except for sweet pellets (18.3%). Highest consumption of sour pellets (81.0%) and pellets with extracts of Chironomids (76.7%), followed by salty (54.1%) and bitter (49.1%) pellets. Higher consumption of 17 of 21 amino acid pellets compared to control (45.4-100% versus 21.8%). Highest consumption of pellets with methionine (100%), alanine (97.2%), glutamine (97.1%), cysteine (97.1%), histidine (91.7%). No deterrent pellets, no difference in holding taste or amino acid pellets in the mouth (1.8-3.9 s for taste pellets, 2.1-4.7 s for amino acid pellets), but longer holding with preferred pellets 16.

• LAB, JUVENILES: females in 140 L aquarium divided into two compartments and either in unlimited contact with males or exposed to selected male stimuli or no contact (control). After 100 days, highest ovarian weight in females with unlimited contact to males (ca 20 g). Lower ovarian weight but still higher compared to control when presented with olfactory or olfactory and auditory (possibly electric) or olfactory and auditory and visual (possibly electric) stimuli (ca 11 g versus 5.3 g control). Lowest ovarian weight when presented with visual and auditory stimuli (5.3-7.3 g in different replicates versus 5.3 g control). Similar results for gonadosomatic index with maximum value under unlimited access (7.3%), medium value at missing tactile stimuli (ca 4%). Results indicate minor importance of visual and auditory stimuli and higher importance of olfactory and tactile stimuli probably displaying additive effect. Because no tactile contact between sexes was observed, difference between group with unlimited contact and missing tactile stimuli could also be due to decreased water quality in groups with missing tactile contact which were presented with waste water from males 72.
• LAB: JUVENILES in 70 L aquarium divided into two compartments and either unlimited contact with opposite sex or exposed to selected male or female stimuli. Males and females were either intact or made anosmic by thermocauterisation of the olfactory epithelium. After 98 days, among males, highest gonadosomatic index (GSI: 0.7%) and seminal vesicle somatic index (SVSI: 0.2%) when alone in fresh water and when exposed to holding water from non-interacting intact females and males (GSI: 0.5%, SVSI: 0.2%). Lowest GSI (0.2%) and SVSI (0.1%) in anosmic males in fresh water and in intact males in direct contact with intact females (GSI: 0.3%, SVSI: 0.1%). Results indicate importance of olfactory sense in males (also by stimuli from other males) and decrease of GSI when in contact with females.
  Among females, highest GSI when in direct contact with intact males (1.5%) or with anosmic males (1.5%). Lowest GSI when alone in fresh water – either intact (0.4%) or anosmic (0.6%). Results indicate importance of olfactory sense in females and increase in GSI when in contact with males 71.
• LAB, JUVENILES: olfactory lobe was third-largest part of the brain (9% volume ratio versus 25% cerebellum, 16% optic tectum, 10% electric-sensitive lateral line lobe) 74.
• LAB: ADULTS in 140 L aquarium divided into two compartments and either unlimited contact with males or exposed to selected male or female stimuli. Females were either intact or made anosmic by thermocauterisation of the olfactory epithelium and injected with pimozide and luteinising hormone-releasing hormone analogue (LHRHa) to induce ovulation. After 33 days, no difference in ovarian growth (36.6-38.3 g) and gonadosomatic index (9.3-10.4%) between intact and anosmic females kept without contact to males. Higher ovarian growth (53.1 g versus 44.8 g) and higher gonadosomatic index (13.1% versus 10.7%) in intact compared to anosmic females kept together with males. Also higher ovarian growth (47 g versus 38.3 g) and higher gonadosomatic index (11% versus 9.3%) in intact compared to anosmic females exposed to holding water from intact females and males. No difference in ovarian growth (38.3-44.8 g) or gonadosomatic index (10.4-10.7%) between anosmic females in contact with males and kept alone. Results indicate importance of olfactory stimuli and non-importance of tactile stimuli for ovarian growth in ADULTS. Larger diameter of oocytes (912 versus 871 µm) and higher relative fecundity in intact compared to anosmic females kept together with males (132.3 versus 100.7 eggs/g body weight) indicate that higher gonadosomatic index is linked to both, larger oocytes and higher number of oocytes 75.
• For importance of olfaction and...
  ...risk perception → D32,
  ...risk perception and coping styles → D12.

• LAB: FRY in 200 x 90 x 90 mm glass aquaria had barbels cauterised and were presented with 50 large Daphnia. During 30 min, lower mean percentage Daphnia consumed compared to control (17.7% versus 41.0%). Result indicates importance of mechanical and possibly chemo-receptive sensing for foraging 14.
• LAB: JUVENILES reacted to approach of experimenter with foraging behaviour and touching the bottom of the tank with the four pairs of large barbels 19.
• For importance of touch and ovarian growth → D31.

• LAB, JUVENILES: electric lateral line lobe was third-largest part of the brain (10% volume ratio versus 25% cerebellum, 16% optic tectum, 9% olfactory lobe) 74.

• LAB: JUVENILES in 100 x 50 x 60 cm tanks (water depth 40 cm) divided into two compartments with shelter in one compartment, feeding station above dipole (pair of vertical carbon electrodes) in other compartment. Field strength of dipole (2.1 mV_p-p/cm) matched that of bulldog Marcusenius macrolepidotus. Stimulus (pulse train of 60 s with pulse rate 30 Hz) was presented, and JUVENILES received reward when immediately left shelter to come to feeding station. After successful trial, stimulus amplitude decreased; if JUVENILES did not leave within 5 s, stimulus amplitude increased. JUVENILES were sensitive to three waveforms (monopolar square-wave, monopolar sine-wave, bipolar sine-wave) with thresholds increasing with decreasing pulse duration. Lowest perception at 4 ms of 13 µV_p-p/cm (monopolar square-wave) 82.

• LAB: electrical sensing spectrum equals detection of electrically discharging M. macrolepidotus of 10.3 cm standard length at maximum of ca 40 cm distance to one of 27.5 standard length at maximum of 150 cm distance. Individuals also detected electrical organ discharges of mormyrid fish species (Hippopotamyrus sp. nov., Hippopotamyrus ansorgii, Mormyrus lacerda) 82.
• For importance of electrical sensing and...
  ...ovarian growth → D31.
  ...interactions with conspecifics → D33.

• LAB: groups of 10 JUVENILES in 70 L aquaria with RAS or flow-through filtration were exposed to 15 mL skin extract (skin diluted in water) from male conspecific. Higher amount of swimming compared to control, but no difference in escape attempts and no effect of filtration type. Results indicate perception not as alarm but as foraging cue possibly due to low concentration of extract. During 14 days, higher amount of swimming in flow-through compared to RAS filtration. No difference in feed intake, growth, and bite wounds 83.
• For signalling stress and coping styles → D12.

• WILD: individuals discharged one-phase waves of >200 ms duration: lake Chamo, Ethiopia 13.
• LAB: JUVENILES discharged monophasic pulses of 5-260 ms duration. Amplitude difficult to measure, because depends on distance to electrodes, and JUVENILES only discharged while moving. Electric field strength at 10 cm distance estimated at 150 µV/cm. Discharges probably originate from electric organ 21.

• LAB: JUVENILES only discharged in aggressive encounters with conspecifics, not when alone or mechanically provoked. Incidences of electrical discharges:
  a) subordinate individual was attacked by dominant,
  b) both individuals were starved, differed in size, were even more confined (especially frequent discharges, series – instead of single – pulses).
  Absence of electrical discharges:
  a) (what might have been) courting,
  b) feeding, catching prey 21.

• WILD: individuals (59.5-70 cm, 2.3-3.7 kg) were observed in aggregations during night and day, seldomly solitary: lake Ngezi, Zimbabwe 7.
• For shoals and feeding → D34.

• WILD: range from 14.4 catch per unit effort at ca 2-4 m depth across day and night to 1.3 catch per unit effort in 15-40 m profundal zone probably due to high availability of food in shallower zone: lake Sibaya, South Africa 9.
• WILD: ca 1 IND/125 m shoreline, range 0-20 IND/125 m: lake Sibaya, South Africa 11.
• For density in spawning aggregations → D1.

• Direct effect: from 10-22 IND/L on:
  • FARM: 3 day-old FRY were stocked in earthen ponds of 100-150 m² and 0.8 m depth surrounded with 0.8 m wall of aluminium roof plates or unprotected at low (≤50 IND/m²), medium (51-100 IND/m²), or high density (101-200 IND/m²) and for ≤50 or ≥51 days. Higher survival in protected than unprotected ponds (38.7% versus 6%). No difference in survival between short and long rearing duration (33.2-39.9%). Higher survival under low and medium compared to high density (53.5% versus 40.4% versus 14.3%) 81.
  • LAB: FRY in 30 L glass aquaria at densities of 150, 300, or 600 IND/tank. At first observation on day 6, higher cannibalism rate the higher the density (ca 2% at 600 IND/L versus 0.5% at 300 IND/L versus 0.2% at 150 IND/L). Difference decreased with increasing time to about 0.5% in all densities at end of observation period on day 47. In general, increasing browsing (ca 55% at 150 IND/tank versus 78% at 600 IND/tank) and decreasing resting activity with increasing density (ca 40% at 150 IND/tank versus 18% at 600 IND/tank) 14.
  • LAB: FRY in 12 L tanks (12 x 30 x 23 cm) at densities of 5, 13, 22, or 30 IND/L and either fed three (08:00, 12:00, 16:00 h) or six times a day (08:00, 10:00, 12:00, 14:00, 16:00, 18:00 h). After 30 days, no difference in cumulative mortality rate (54%). Differences in the frequency of changing from one behaviour to resting, swimming or browsing: less switching to resting under highest density compared to the two lowest densities (ca 6 times versus 10.5-11 times); less switching to swimming behaviour under 22 IND/L compared to 5 IND/L (ca 7 times versus 11.5 times); more switching to browsing under 13 IND/L compared to 30 IND/L (ca 11 times versus 6.5 times). Results indicate fewer switching of behaviour under higher densities.
    Higher frequency of resting (56% versus 24-35%) and lower frequency of swimming under 5 IND/L compared to 22 and 30 IND/L (ca 35% versus 55-68%), 13 IND/L in between (resting ca 45%, swimming ca 40%). No difference in browsing 26.
• No effect:
  • LAB, FRY: no difference in survival in stocking densities of 50, 100, 150, 200, or 250 IND/L and with or without shelter (80-96.8%; for details on study → D26) 23.
  • LAB, FRY: no difference in survival in stocking densities of 50, 100, or 150 IND/L (ca 90.5-93.7%; for details on study → D26), but increasing non-cannibalistic mortality with decreasing density (7.5% at 50 IND/L, 4.7% at 100 IND/L, 2.8% at 150 IND/L) probably due to aggressive territoriality 24.
  • LAB: JUVENILES (7.5 g) in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 500 IND/m³, 1,125 IND/m³, 1,750 IND/m³, 2,375 IND/m³, or 3,000 IND/m³ that correspond to 50 kg/m³, 112.5 kg/m³, 175 kg/m³, 237.5 kg/m³, and 300 kg/m³ at the end of the first growth period (at ca 100 g weight). After 49 days, no difference in plasma cortisol (ca 25-65 nmol/L). Slight decrease in glucose with increasing density until 2,375 IND/m³ (ca 2 mmol/L), thereafter increase.
    Tendency of increasing plasma lactate with increasing density (ca 3 mmol/L at 3,000 IND/m³). Increasing escape attempts with increasing density (0.01-0.09 attempts/IND/h at 500-1,750 IND/m³ versus 0.5-0.6 at 2,375-3,000 IND/m³), but decreasing over observation time. Increase in air breathing acts with increasing density (21.3 acts/IND/h at 500 IND/m³ versus 109.3 acts/IND/h at 3,000 IND/m³). Decrease in skin lesions with increasing density (5.1 at 500 IND/m³ versus 2.0 at 3,000 IND/m³). No difference in swimming activity (79.8-97.5%), stereotypical behaviour (continuous and compulsive swimming under a fixed, repetitive pattern for at least 10 s; 0-0.02 times/IND/h), and aggressive acts (0-4 acts/IND/h) 33.
  • LAB: JUVENILES (102.1 g) in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 67.7 IND/m³, 133.3 IND/m³, 267.7 IND/m³, 533.3 IND/m³, or 1,067.7 IND/m³ that correspond to 20 kg/m³, 40 kg/m³, 80 kg/m³, 160 kg/m³, and 320 kg/m³ at the end of the second growth period (at ca 300 g weight). After 28 days, no difference in plasma cortisol (177.5-373.6 nmol/L), glucose (2.8-3.8 mmol/L), and lactate levels (3.8-4.7 mmol/L). Increasing swimming activity (50.4% at 67.7 IND/m³ versus 93.9% at 1,067.7 IND/m³) and increasing air breathing acts (43 acts/IND/h at 67.7 IND/m³ versus 80 acts/IND/h at 1,067.7 IND/m³) with increasing density. Decrease in skin lesions (1.4 at 67.7 IND/m³ versus 1.0 at 1,067.7 IND/m³). No difference in escape attempts (0-0.5 attempts/IND/h), aggressive acts (0-8 acts/IND/h), and stereotypical behaviour (continuous and compulsive swimming under a fixed, repetitive pattern for at least 10 s; 0 times/IND/h).
    Stress test for half of sampled individuals by keeping them in nets outside water for 10 min and placing them in 30 L tanks (5 IND/tank) for 1 h. Surprisingly, lower cortisol levels than in unstressed individuals (134.7-201.7 nmol/L versus 177.5-373.6 nmol/L without stress), no difference between densities. Decrease in number of skin lesions with increasing density (5.2 at 67.7 IND/m³ versus 1.1 at 1,067.7 IND/m³).
    Experiment 2: ADULTS (1,044.6 g) in 200 L glass aquaria of a recirculating aquaculture system at densities of 67 IND/m³, 155 IND/m³, 244 IND/m³, or 333 IND/m³ that correspond to 100 kg/m³, 233 kg/m³, 366 kg/m³, and 500 kg/m³ at the end of the second growth period (at ca 1,500 g weight). After 28 days, no difference in plasma cortisol (251.4-383.3 nmol/L), glucose (2.7-3.7 mmol/L), and lactate levels (2.8-3.2 mmol/L) between densities. No difference in swimming activity (42.1-48.1%), air breathing acts (50-73 acts/IND/h), escape attempts (0.6-1 attempts/IND/h), aggressive acts (0-13 acts/IND/h), stereotypical behaviour (0), and skin lesions (1.4-2.1).
    Stress test for half of sampled individuals by keeping them in nets outside water for 10 min and placing them in 30 L tanks (5 IND/tank) for 1 h. Surprisingly, lower cortisol levels than in unstressed individuals (80.2-151.5 nmol/L versus 251.4-383.3 nmol/L without stress), no difference between densities. No difference in glucose (6.1-6.7 mmol/L) and lactate levels (3.6-4.9 mmol/L) and number of skin lesions (2.1-2.4) between densities.
    Results of both experiments indicate importance of other – still to be identified – factors besides stocking density affecting welfare 34.
  • LAB: JUVENILES in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 100 kg/m³ (8 IND/tank) or 500 kg/m³ (40 IND/tank) and either composed of proactive or reactive individuals (→ D12) or a mix. Higher aggression under lower density (20 acts/IND/h versus 4 acts/IND/h). No difference in number of lesions (1.1-1.6), air breathing behaviour (37.5-40.7 acts/IND/h), swimming time (59.8-71.5%), resting time (28.5-40.2%), and escape behaviour (0.1-0.2 acts/IND/h).
    Lower latency of feeding (0 versus 0.2 s) and higher total feeding speed (1.5 versus 0.3 g/s) under higher density, but no difference in total feeding time between densities (198-209 s). Under low density, higher number of skin lesions in proactive and reactive than mixed groups (ca 2.2 versus 0.5), but no difference under high density (ca 0.5-0.9) 29.

• Inverse relation: from 10-100 IND/L on:
  • FARM: 3 day-old FRY were stocked in earthen ponds of 100-150 m² and 0.8 m depth surrounded with 0.8 m wall of aluminium roof plates or unprotected at low (≤50 IND/m²), medium (51-100 IND/m²), or high density (101-200 IND/m²) and for ≤50 or ≥51 days. Higher weight at harvest (26.7 g versus 3.1 g) and higher growth rate (20.1%/d versus 17.7%/d) but lower biomass (359 kg/ha versus 735 kg/ha) in unprotected than protected ponds. For survival → D2.
    In protected ponds, higher growth rate (18.7% versus 9.8%) but lower weight at harvest (2.3 g versus 7.2 g) and lower biomass (652 kg/ha versus 1,149 kg/ha) after short compared to long rearing duration probably due to small group of larger FRY (0-3% of 8-12 g versus 97% of 0.5-3 g) at end of short rearing period that could cannibalise on siblings. No difference in weight at harvest, but higher biomass under low compared to medium and high density (1,203 kg/h versus 689 kg/ha versus 540 kg/ha), lower growth rate under low compared to medium and high density (11.6%/d versus 18.4%/d versus 15.9%/d) 81.
  • LAB: FRY in 30 L glass aquaria at densities of 150, 300, or 600 IND/tank. After 26 days, higher growth the lower the density (ca 55 mm under 150 IND/tank versus 47 mm under 300 IND/tank versus 40 mm under 600 IND/tank) 14.
  • LAB: 48 h-old FRY in 10 L cylindrical plastic tanks with lids at densities of 50, 100, 150, 200, or 250 IND/L. Half of tanks contained shelters in form of cylindrical 4 mm inert plastic mesh. After 6-7 days, peak of specific growth rate, thereafter decrease to low point at 10-14 days (end of observation period). Faster decrease the higher the stocking density.
    Second experiment with 48 h-old FRY in 10 L cylindrical plastic tanks with lids at densities of 25 IND/L over 20 days or 50 IND/L over 16 days. Higher weight under 25 IND/L than 50 IND/L (ca 125 mg after 20 days versus ca 50 mg after 16 days) 23.
  • LAB: first-feeding FRY in 1 m diameter rearing tanks at densities of 50, 100, or 150 IND/L. After 35 days, highest weight (ca 350 mg versus ca 180-230 mg) and highest specific growth rate (ca 0.14 versus 0.13) under lowest density 24.
  • LAB: FRY in 12 L tanks (12 x 30 x 23 cm) at densities of 5, 13, 22, or 30 IND/L and either fed three (08:00, 12:00, 16:00 h) or six times a day (08:00, 10:00, 12:00, 14:00, 16:00, 18:00 h). After 30 days, higher growth (ca 60 mm versus ca 45-50 mm) and higher growth rate under lowest than higher densities (1.9 mm/d versus 1.4 mm/d) 26.
• No effect:
  • LAB: JUVENILES in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 500 IND/m³, 1,125 IND/m³, 1,750 IND/m³, 2,375 IND/m³, or 3,000 IND/m³ that correspond to 50 kg/m³, 112.5 kg/m³, 175 kg/m³, 237.5 kg/m³, and 300 kg/m³ at the end of the first growth period (at ca 100 g weight). After 49 days, no difference in final weight (86.7-93.7 g), specific growth rate (4.9-5.1), and FOOD CONVERSION RATIO (0.6-0.7) 33.
  • LAB: JUVENILES (102.1 g) in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 67.7 IND/m³, 133.3 IND/m³, 267.7 IND/m³, 533.3 IND/m³, or 1,067.7 IND/m³ that correspond to 20 kg/m³, 40 kg/m³, 80 kg/m³, 160 kg/m³, and 320 kg/m³ at the end of the second growth period (at ca 300 g weight). After 28 days, increase in feed intake with increasing density (3.6 g/d at 67.7 IND/m³ versus 4.7 g/d at 1,067.7 IND/m³). No differences in specific growth rate (1.9-2.3 %/d) and FOOD CONVERSION RATIO (0.7-0.8).
    Experiment 2: ADULTS (1,044.6 g) in 200 L glass aquaria of a recirculating aquaculture system at densities of 67 IND/m³, 155 IND/m³, 244 IND/m³, or 333 IND/m³ that correspond to 100 kg/m³, 233 kg/m³, 366 kg/m³, and 500 kg/m³ at the end of the second growth period (at ca 1,500 g weight). After 28 days, no difference in feed intake (11.7-12.3 g/d), specific growth rate (1.2-1.3%/d), FOOD CONVERSION RATIO (0.8-0.9) 34.
  • LAB: JUVENILES in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 100 kg/m³ (8 IND/tank) or 500 kg/m³ (40 IND/tank) and either composed of proactive or reactive individuals (→ D12) or a mix. No difference in food intake (16.2-17.2 g/kg^0.8/d) and specific growth rate (2.9%/d) between densities. Tendency of higher FOOD CONVERSION RATIO under high compared to low density (0.71 versus 0.66). Under low density, higher specific growth rate in reactive than proactive than mixed groups (ca 3.4% versus 3% versus 2.4%), but no difference under high density (2.8-2.9%) 29.
  • For stocking density in polyculture → D35.

• Hierarchy and group size: in small groups, individuals establish linear dominance hierarchies:
  • Observations LAB, FRY: in groups of five, one became dominant (→ D6) 22.
  • Observations LAB, FINGERLINGS: in pairs, one became dominant 16.
  • Observations LAB, JUVENILES: in pairs, one became dominant 21, in groups of four, one became dominant 19.
• Establishing hierarchy:
  • LAB: FRY in 200 L tanks (87 x 58 x 46 cm) at density of 36 IND/tank and sorted into four groups: homogeneous low weight (L, mean 83 g, range 66.5-94.9 g), homogeneous medium weight (M, 140 g, 125.4-154.9 g), homogeneous heavy weight (H, 198 g, 184.5-214.1 g), heterogeneous group with one third of other groups (HET, 139.7 g, 65.2-214.3 g). After 34 days, among similar-sized FRY in L, M, and HET groups, no difference in number of skin lesions and individuals bitten, but in heavy FRY, higher number of skin lesions (2.0 versus 1.4) and higher number of individuals bitten (67.9% versus 58.3%) in HET than H. Results indicate that either social hierarchies are not established or that size is not important in establishing hierarchies as well as that low weight individuals are not automatically subordinate 32.
  • LAB: two FINGERLINGS differing in length by 1-2 cm were transferred from holding tank to 3 L aquaria. In the majority of aquaria, the larger was very aggressive, inflicted so bad injuries (bitten-off caudal fin or part of caudal peduncle) that some of the smaller FINGERLINGS died and had to be replaced. Aggression decreased after about 10 days 16.
  • LAB: pairs of JUVENILES in 20 x 20 x 30 compartment in aquarium began to fight shortly after being introduced, interspersed with pauses 21.
  • LAB, JUVENILES: 3-6 h after transfer to experimental tank, JUVENILES began to move and soon after began to fight. Established hierarchy within the course of the first day 19.

• Features of dominance:
  • WILD, ADULTS: aggressors in prenuptial aggressive encounters between males directed barbels forward, erected fins: lake Sibaya, South Africa 11.
  • LAB: dominant JUVENILES were first to feed, chased subordinate away, followed the ones who managed to get feed 19.
• Hierarchy and size:
  • LAB, FRY: in groups of five, larger individuals visited colour plates with reward first while smaller individuals waited in aquarium corners (→ D6) 22.
  • LAB, FINGERLINGS: in pairs, in the majority of aquaria, the larger (1-2 cm) became dominant, chased subordinate 16.
  • For social hierarchy and growth → D36.

• Features of subordination:
  • WILD: submissive ADULTS in prenuptial aggressive encounters between males flattened barbels and fins: lake Sibaya, South Africa 11.
  • LAB: subordinate JUVENILES tried to escape attacks by jumping out of the aquarium or swimming against walls or corners causing injuries, rested on bottom with maximum possible distance to dominant individual 19.
• Hierarchy and stress: no data found yet.

• LAB: 7 day-old FRY in 250 m³ aquarium with Myriophyllum spicatum and sand or in an enamel tray fed on dead C. gariepinus FRY 11.
• LAB: FRY in 600 x 400 x 400 mm glass aquaria (250 mm depth, 30 L) cannibalised when actively came in contact with siblings whereas territorial behaviour was initiated by being invaded by sibling (mostly via head-on-head contact). First incidences of cannibalism at 8 mm TOTAL LENGTH (3.5 days after start of exogenous feeding) of type I (where the head remains). Until 45 mm TOTAL LENGTH, homogeneous size distribution and thus prey 80% (range 52.9-142.9%) of size of predator. From 45 mm TOTAL LENGTH on, more heterogeneous size distribution and increase in type II cannibalism (where the prey is swallowed whole) where prey is 40% (range 37.7-53.1%) of size of predator. Differences in size are probably the result not the cause of cannibalism, as cannibals benefit from nutritional advantage and increase in growth 14.
• LAB: among 136 deaths of FRY during 14-30 days observation period under stocking densities of 50, 100, 150, 200, or 250 IND/L (for details on the study → D26), 5 attributed to cannibalism type I (where heads remain). Occurred at beginning of air breathing (ca days 14-16) where flight response was hindered by increased buoyancy and when weaned from Artemia to powdered diet 23.
• LAB: 2-3.6% of deaths under densities of 50, 100, or 150 IND/L (for details on study → D26) due to cannibalism type I (where heads remain) with first peak at beginning of air breathing and second peak when weaning from 790 µm to 1,300 µm particles. No type II cannibalism 24.
• For cannibalism and...
  ...shelter → D10,
  ...food competition → D37,
  ...PHOTOPERIOD → D38,
  ...stocking density → D2.

• For cooperation and pack hunting → D34.

• LAB: FRY in 12 L tanks (12 x 30 x 23 cm) at densities of 5, 13, 22, or 30 IND/L and either fed three (08:00, 12:00, 16:00 h) or six times a day (08:00, 10:00, 12:00, 14:00, 16:00, 18:00 h). During first weeks, no difference in number of aggressive contacts between densities but highest rate of aggression under low than other densities (70%). During 10 s of swimming, higher number of aggressive contacts under low compared to two highest densities (aggressive during 77.5% of contacts versus 51.8-52.2%), but lower frequency of swimming under low compared to highest densities (→ D2). Difference between densities decreased as FRY grew beyond 25 mm 26.

• LAB: first of eight rays in pectoral fins are much larger and can be erected when individuals feel disturbed. Humans were injured by pectoral stings when handling C. gariepinus in supermarkets. Endured envenomation with severe pain, oedema, erythema. Wounds healed slowly despite oral antibiotics 84.

• Size-matched pairs:
  • LAB, JUVENILES: attacks, bites, chases 21.
• Non-matched pairs:
  • LAB, FINGERLINGS: chases, grasps, bites (biting off caudal fin or part of caudal peduncle) 16.
  • LAB, JUVENILES: more severe aggression as in size-matched pairs, constant fighting 21.
  • LAB, JUVENILES: blows, chases, bites on body, fins, barbels 19.

• Managing self-feeder:
  • LAB: JUVENILES learned how to manage a self-feeder (→ D39 D25) 27 18.

• Feed efficiency versus inefficiency after risk perception:
  • LAB: individual JUVENILES in 30 L aquaria exposed to 5 mL skin extract (skin diluted in water) from male conspecific. No difference in swimming time between JUVENILES exposed to skin extract and control. JUVENILES differed in residual feed intake (RFI), that is the difference between the feed consumed and that predicted by a regression model. JUVENILES with low RFI were feed efficient and decreased swimming in response to skin extract. JUVENILES with high RFI were feed inefficient and increased swimming time. No freezing or immobility. Results indicate a difference in coping styles masking reaction to conspecific skin extract 28.
• Proactivity versus reactivity:
  • LAB: individual JUVENILES in tanks from which water was drained until 1.5 cm depth. Based on their behaviour during 5 min, individuals were categorised as proactive (the 25% most active individuals that attempted escape [mean 7.8 escape attempts]) or reactive (the 25% least active without escape attempts) or intermediate. Performed in experiment (→ D2) and were exposed to escape test again. Displayed different behaviour given the group composition in the experiment: When in group with proactive or mixed individuals, decrease in escape attempts (ca 2.2 in first test versus 7.5 in second test in proactive group; 1 versus 2.5 in mixed group) and moving behaviour compared to first escape test (ca 31% versus 47% in proactive group; 20% versus 28% in mixed group). Tendency of increase in escape attempts (1 versus 0) and moving activity (ca 15% versus 10%) in reactive individuals. Results indicate only partial consistency over time and instead the influence of other factors on behaviour probably because variation benefits group interest.
    Lower feed intake (15.4 g/kg^0.8/d versus 17.3 g/kg^0.8/d), specific growth rate (2.6%/d versus 3%/d), and tendency of higher FOOD CONVERSION RATIO (0.71 versus 0.66-0.68) in reactive than proactive or mixed groups. Lower gross energy intake in reactive then proactive or mixed groups (315 kJ/kg^0.8/d versus 354 kJ/kg^0.8/d) and consequently also tendency of lower faecal energy loss (36.7 kJ/kg^0.8/d versus 41.9-43.2 kJ/kg^0.8/d), total digestible energy (278 kJ/kg^0.8/d versus 310-312 kJ/kg^0.8/d), metabolisable energy (263 kJ/kg^0.8/d versus 295-296 kJ/kg^0.8/d), and retained energy (171 kJ/kg^0.8/d versus 191-200 kJ/kg^0.8/d). No difference in number of lesions (1.2-1.6), in aggression (8-16.5 acts/IND/h), air breathing behaviour (36.3-43 acts/IND/h), swimming time (58.1-73.2%), resting time (26.9-42%), and escape behaviour (0.1-0.3 acts/IND/h). No difference in latency of feeding (0.1 s) and total feeding speed (0.9-1 g/s), but higher total feeding time in mixed compared to proactive or reactive groups (234 s versus 175-200 s) 29.
• For coping styles and handling stress → D14.

^{In the structure of menu item 16 and the definition of "SHYNESS-BOLDNESS", we follow 85.}

• Neophobia:
  • LAB: two FINGERLINGS differing in length by 1-2 cm were transferred from holding tank to 3 L aquaria. Were agitated and frightened when experimenter approached, tried to hide behind heater. In first days, food consumption only when experimenter went away. Did not go after pellet when sank on bottom in (even short) distance, so pellets were delivered on head 16.
  • LAB: JUVENILES were transferred from holding to 235 L experimental tanks (150 x 65 x 30 cm) and fed 3 h later. No feeding in 11 of 16 tanks, maximum 2 pellets in remaining tanks. Movement of experimenter increased anxiety. At 24 h after transfer (next observation time), movement of experimenter caused foraging behaviour, feeding amount increased 19.

^{In the structure of menu item 16 and the definition of "EXPLORATION-AVOIDANCE", we follow 85.}

• For aggressiveness and...
  ...establishing hierarchy → D42,
  ...dominance → D11,
  ...subordination → D43,
  ...stocking density → D44,
  ...size-grading → D45.

^{In the structure of menu item 16 and the definition of "AGGRESSIVENESS", we follow 85.}

• For fear and neophobia → D13.

• For stereotypical activities and stocking density → D2.

• Air exposure:
  • LAB: individual JUVENILES (58.9 g) in 40 L aquaria (30 x 35 x 40) were fed manually with batches of 5 pellets each until apparent satiation. After 32 days, stress test: keeping individuals in net outside water for 1 h. Similar second experiment with larger JUVENILES (75.5 g) and duration of 24 days. Higher glucose levels (median ca 6-6.5 mmol/L versus 2 mmol/L) in both experiments in stressed compared to control JUVENILES. Trend of higher lactate (median ca 4 mmol/L in experiment 1, median 2.5 mmol/L in experiment 2 versus 2 mmol/L) and cortisol levels in stressed compared to control JUVENILES (median ca 60 ng/mL versus 40 ng/mL). Regression model indicated that glucose and cortisol levels explained minor part of differences in feed efficiency in that highly stressed individuals were less efficient feeders. Results also indicate that individual differences in stress response are independent of social interactions and probably genetically determined 73.
  • LAB: JUVENILES (7.5 g) in 120 L glass aquaria (90 x 45 x 45 cm) of a recirculating aquaculture system at densities of 500 IND/m³, 1,125 IND/m³, 1,750 IND/m³, 2,375 IND/m³, or 3,000 IND/m³ that correspond to 50 kg/m³, 112.5 kg/m³, 175 kg/m³, 237.5 kg/m³, and 300 kg/m³ at the end of the first growth period (at ca 100 g weight). After 49 days, stress test for half of sampled individuals by keeping them in nets outside water for 10 min and placing them in 30 L tanks (10 IND/tank) for 1 h. Increasing plasma glucose (ca 5-6 mmol/L versus 2.2-3.8 mmol/L without stress) and lactate levels (ca 4-5.5 mmol/L versus 2.5-3 mmol/L without stress) compared to individuals without stress test independent of densities. No difference in cortisol levels in stressed compared to unstressed individuals at 500 IND/m³ versus 3,000 IND/m³ (ca 60-65 nmol/L) probably due to dampening of ACTH during chronic stress. Higher number of skin lesions than in unstressed individuals and increase with increasing density (ca 8.2 at 500 IND/m³ versus 11.5 at 3,000 IND/m³) 33.
• Collection and weighing:
  • LAB, FRY: increased number of deaths on days after weighing 23.

• FARM: individuals (1-1.5 kg) in 4,000 or 12,000 L concrete tanks of a recirculating aquaculture system at density up to 250 kg/m³. Before transport fasted for 24 h, then transferred to flow-through tanks for 48 h of fasting at density 500 IND/tank. To simulate transport, water was drained for 40 cm, individuals were transferred to empty tank for weighing, dropped for 1 m into partially filled 120 x 100 x 79 cm plastic tank resembling usual transport tanks at density of 500 kg/m³. Covered tanks were loaded onto truck, transported for 3 h. After return, tanks were filled with fresh water to decrease density to 250 kg/m³. No difference in skin lesions compared to control individuals which were also transferred to plastic tanks and fasted for 72 h (ca 12 lesions/IND) but increase in lesions compared to before handling (12 lesions/IND versus 8 lesions/IND). Increase in plasma cortisol after preparation for transport compared to before handling (35 ng/mL versus <10 ng/mL) and during transport compared to control (50 ng/mL versus 30 ng/mL). At 6 h after transport, cortisol had returned to basal level in control individuals (indicating stress probably involved with handling, transfer to unfamiliar and increased density in transport tanks), but remained increased in transported individuals at least until next observation time at 24 h after transport, returned to basal levels at next observation time at 48 h after transport. No difference in plasma glucose in transported and control individuals compared to before handling (ca 5.8-6.5 mmol/L), but decrease at 48 h after transport in transported and control individuals compared to basal levels (ca 3.7-4.2 mmol/L versus 5.8-6.5 mmol/L) 86.

• Size-grading:
  • LAB: FRY in 120 L tanks (90 x 45 x 45 cm) at density of 36 IND/tank and sorted into four groups: homogeneous low weight (L, mean 83 g, range 66.5-94.9 g), homogeneous medium weight (M, 140 g, 125.4-154.9 g), homogeneous heavy weight (H, 198 g, 184.5-214.1 g), heterogeneous group with one third of other groups (HET, 139.7 g, 65.2-214.3 g). After 27 days, before feeding, lower resting (48.7% versus 53.7-56.2%) and higher swimming activity (51.3% versus 43.8-46.3%) in H compared to L and M. Higher waiting-in-feeding-area activity compared to L and M (26.8% versus 17.1-1.5%), HET in between (22.9%). After feeding, lower resting (17.2-20% versus 29.7%) and higher swimming activity (80-82.8% versus 70.3%), especially waiting-in-feeding area (51.5% in H versus 42.8-43.9% in M and HET versus 26.9%), in H, M, and HET compared to L. Higher feeding time the larger the weight in the group (3.2 min in L versus 4.7 min in M versus 5.5 min in H), HET in between (4.7 min in HET). Higher feeding time in M, H, and HET compared to L (29.4-29.7 g/min versus 23.8 g/min) 30.
  • LAB: FRY in 200 L tanks (87 x 58 x 46 cm) at density of 36 IND/tank and sorted into four groups: homogeneous low weight (L, mean 83 g, range 66.5-94.9 g), homogeneous medium weight (M, 140 g, 125.4-154.9 g), homogeneous heavy weight (H, 198 g, 184.5-214.1 g), heterogeneous group with one third of other groups (HET, 139.7 g, 65.2-214.3 g). After 34 days, fewer skin lesions in H than L (1.4 versus 2.0) and fewer individuals bitten in H than all other groups (58.3% versus 69.7- 74.1%). Number of skin lesions decreased over time in M, number of individuals bitten decreased in M and H. Results indicate possible advantage of size-grading in M and H individuals.
    Among similar-sized FRY, no difference in cortisol in all groups, but in medium FRY, higher cortisol in HET than M (73.5 ng/mL versus 45.7 ng/mL); no difference in glucose in all groups. Result indicates that size-grading does not lower chronic stress.
    Stress test: 10 FRY each from the different groups (chronic stress) versus 10 each from groups that were additionally exposed to air for 1 h in net (acute stress). Increase in plasma cortisol in all groups (ca 140-160 ng/mL versus 40 ng/mL) except L, increase in plasma glucose (ca 5.5-7 mmol/L versus 3.5 mmol/L). Among similar-sized acutely stressed FRY, no difference in cortisol in all groups; no difference in glucose in all groups, but in low FRY, lower glucose in HET than L (5.1 mmol/L versus 7.1 mmol/L) 32.

• Stunning rules: to minimise pain reactions and enhance welfare before slaughter:
  1. induce insensibility as fast as possible,
  2. prevent recovery from stunning,
  3. monitor effectiveness (observations, neurophysiological measurements) 87.

• Comparison of stunning methods:
  • LAB: Experiment 1: individuals (1,083 g) were implanted with electrodes for EEG and ECG and shot with 16 mm captive needle which inserted compressed air into brain at shooting pressure of 8 bar. No brain activity for 11 of 21 individuals after 11 s. For the remaining 10 individuals, brain activity was maintained until end of observation 10 min after stunning but probably without perception of pain. Heartbeat of 71 beats/min before stunning, fibrillation afterwards in nine of 21 individuals, irregular heartbeats in the rest.
    Experiment 2: individuals (1,688 g) were implanted with electrodes for EEG and ECG and stunned by head-only electrical stunning for 1 s at 350 V (50 Hz AC), thereafter captive needle pistol. No brain activity in 16 of 21 individuals after 16 s. For the remaining five individuals, brain activity was maintained until end of observation 10 min after stunning but probably without perception of pain. Heartbeat of 58 beats/min before stunning, fibrillation afterwards in three of 21 individuals, irregular hearbeats in the rest.
    Based on Experiment 1 and 2, authors calculated chance of effective stun as 93-100% when done correctly.
    Experiment 3: individuals (1,793 g) moved freely in tank, were stunned by captive needle, placed in water. No normal swimming possible, bent into U-shape, bumped into tank walls. Motionless after 38 s.
    Experiment 4: individuals (1,704 g) moved freely in tank, were stunned by captive needle, placed in ice water. No movement in five of seven individuals. The remaining two displayed clonic cramps for 3 and 63 s 88.
  • LAB: Experiment 1: individuals (1,536 g) were implanted with electrodes for EEG and ECG and electrically stunned individually in a 50 x 20 x 70 cm plexiglass box in the head only for 1 s at 291 Volt (50 Hz, sinusoidal AC), 1.6 A/dm². 24 individuals displayed epileptiform insult: 7 s tonic phase, 12 s clonic phase, 9 s exhaustion. Authors calculated chance of effective stun as 88-100%. Heartbeat of 77 beats/min before stunning, 11 s fibrillation afterwards, increasing, then decreasing heartbeat (69 beats/min at 5 min after stunning), mainly irregular heartbeats. All individuals responded to pain stimuli until last observation at 5 min after stunning.
    Second stunning either at a) 300 Volt, 1.7 A/dm² for 10 s (five individuals), b) 300 Volt, 1.7 A/dm² for 1 s followed by 100 Volt, 0.5 A/dm² for 5 min (three individuals) or 300 Volt, 1.7 A/dm² for 1 s followed by 50 Volt, 0.3 A/dm² for 5 min (three individuals), c) 300 Volt, 1.7 A/dm² for 10 s, decapitation 36 s later (15 individuals). Decreased brain activity. Some individuals regained consciousness at observation time 0.5 min after stunning except for decapitated ones. Results indicate non-effectiveness of head-only stunning except when combined with decapitation and bleeding.
    Experiment 2: individuals (1,738 g) were implanted with electrodes for EEG and ECG and chilled alive in ice water (0.1 °C) for 30 min while being restrained. Loss of response to pain stimuli at median 12.5 min (range 5-20 min), decreased brain activity at median 15 min. Authors calculated chance of effective stun as 87% after 30 min. No change in heartbeat until 30 min after stunning (295 beats/min at beginning versus 300 beats/min after 30 min; tachycardia) probably due to cold restrainer. Nine of 22 individuals moved vigorously when placed in ice water. Ten individuals responded to pain stimuli until last observation time at 30 min.
    Experiment 3: individuals (2,422 g) were placed in ice water (-0.1 °C) for 10 min, allowed to move freely. After swimming for 13 s, 284 s clonic phase, loss of movement after 300 s.
    Results (long time until loss of movement, cramps, tachycardia) indicate that live chilling (Experiment 2 and 3) is a possibly stressful method 89.