The effect of stocking density on growth,survival and agonistic behaviour of African catfish

A high rate of sibling cannibalism is one of the principal obstacles in the rearing of larvae and juveniles of the African catfish Clarias gariepinus. This paper examines the underlying behavioural components of agonistic behaviour under culture conditions. Ten day old catfish larvae were stocked at 30 larvae l^–1 in three tank designs with different surface areas and equal volumes. Stocking densities were 1.2, 0.6 and 0.3 fish cm^–2 bottom surface. Growth did not differ between treatments. The highest mortality over 30 days was recorded at the medium density (0.6 larvae l^–1). Fish stocked at this density showed the highest rate of aggression, while there was no difference in aggression between the highest and the lowest stocking densities. The results indicate that stocking density should be at least as high as 1.2 larvae cm^–2 bottom surface area to obtain high production at best survival rates.     

Fish-management relationships in Israeli commercial fish farming

To analyse the relationships among fish species performance and management procedures, a database was built up with data from 31 fish farms during the period 1976–1987 (1673 observations) and analysed through multivariate statistics (factor analysis). The data include nurseries, grow-out and operational ponds with mono- and polycultures of common carp, Cyprinus carpio, tilapia hybrid, Oreochromis niloticus × O. aureus, silver carp, Hypophthalmichthys molitrix, and mullet, Mugil cephalus. The main conclusions include the following. (1) The highest total yields and best tilapia performances were obtained in polyculture ponds where tilapia was the main species. (2) The best carp performances occurred in grow-out polyculture ponds where carp was the main species. Carp performance was improved in polycultures with mullet and silver carp, irrespective of whether tilapia were present or not. (3) Carp and tilapia yields increased as the nutritional inputs (feed pellets, sorghum pellets, manure), pond size and culture duration increased. The effect of the nutritional input was not linear, but logarithmic. (4) Growth rate of common carp was more affected by total density and stocking size than that of tilapia. Better carp and tilapia growth occurred in grow-out ponds when stocked at large sizes and cultured during short periods, mainly when both species were present. (5) Carp growth varied with the geographical region and size of fish pond, being better in smaller than in larger ponds due to reduced access to natural benthic food in deep ponds.     

Effects of predation by largemouth bass in fish production ponds stocked with Tilapia nilotica

Triplicate ponds (0.07 ha) were stocked with either 0, 29, 86, or 143 largemouth bass fingerlings and 9500 Tilapia nilotica fingerlings (90% males) per hectare. Fish were fed a 32% protein ration for 6 months and then harvested. Average weight of largemouth bass declined as their stocking density increased. Average weight and total biomass of originally stocked tilapia increased with increasing predator stocking density. Total yields of tilapia were not different and averaged 4203, 4355, 4583 and 4230 kg/ha, of which recruits accounted for 28, 29, 26 and 8% for the above respective predator densities. Stocking 143 largemouth bass per hectare was an effective means of decreasing recruitment and increasing yields of larger sized male tilapia.     

Effect of density on survival and growth of cyprinid fish fry

One of the bottlenecks in Israel's cold water ornamental fish industry is the large loss of fish during the post-larval stage. As a first step towards increasing survival rates, the optimal stocking density in earthen ponds for fry of each species should be determined. The results of five consecutive experiments with fry of goldfish (Carassius auratus), common carp and koi (respectively, edible and ornamental morphotypes of Cyprinus carpio) are presented here. The experiments were run in 18 cages of 1 m³ and 0.8 mm mesh size placed in a 0.02 ha earthen pond, stocked at 50,100 and 200 fry per cage. The variables analyzed were growth rate, harvesting weight, biomass and survival. Equations to estimate growth rates under different conditions are given. The three fish types present different growth rates and survival patterns in relation to stocking density and season, which lead to different management implications. Koi can be stocked up to 2 million fry per ha without showing negative density effects. Goldfish should be stocked at low density (500,000–1 million fry per ha) in spring and at higher rate (2 million fry per ha) in summer. Common carp stocking density should be adjusted to obtain the required fish size at harvest.     

Grow-out culture of tropical abalone, Haliotis asinina (Linnaeus) in suspended mesh cages with different shelter surface areas

Thisstudy investigated the effects of shelter surface area (SSA) on the feeding,growth and survival of the donkey-ear abalone, Haliotisasinina reared in mesh cages (0.38×0.38×0.28m) suspended in flow-through tanks (water volume = 6m³). Cages had sections of polyvinylchloride (PVC) thatprovided shelters with surface area of 0.22 m², 0.44m² and 0.66 m².Hatchery-produced abalone with initial shell length of 32 ± 1mm and wet weight of 7.5 g were stocked at 50individuals cage^–1 that corresponded to stocking densities ofca. 227, 113 and 75 abalone m^–2 of SSA. The ratios of sheltersurface area to cage volume (SSA:CV) were 5.5, 11 and 16.5. Abalones wereprovided an excess red seaweed Gracilariopsis bailinae(= Gracilaria heteroclada) at weekly intervals overa 270-day culture period. Feeding rates (18–20% of wet weight), foodconversion ratio (26–27) and percent survival (88–92%) did notdiffer significantly among treatments (p > 0.05). Body size at harvest rangedfrom 56 to 59 mm SL and 52 to 57 g wet body weightwith significant differences between abalone reared at SSA 0.22m² and 0.66 m² (p < 0.05).Abalone reared in cages with 0.66 m² SSA grewsignificantly faster at average daily growth rates of 132 m and188 mg day^–1. Stocking densities of 75–113m^–2 SSA in mesh cages suspended in flow-throughtanks resulted in better growth of abalone fed red seaweed.     

Effect of Stocking Density on Growth and Water Quality for Largemouth Bass Micropterus salmoides Growout in Ponds

Juvenile largemouth bass Micropterus salmoides , trained to accept artificial diets, were stocked into six 0.04-ha ponds at stocking densities of either 6,175 or 12,350 fish/ha. Fish were fed a floating custom-formulated diet, containing 44% protein, once daily to satiation for 12 mo (May 1994–May 1995). At final harvest, the total yield of fish was significantly greater (P < 0.05) and feed conversion ratio (FCR) was significantly lower, for bass stocked at the higher density (4,598 kg/ha and 2.3, respectively) than when stocked at the lower density (2,354 kg/ha and 3.3, respectively). There was no significant difference (P > 0.05) in average weight, length, or survival of bass stocked at the two densities. Averaged over the study period, there were no significant differences (P > 0.05) in total ammonia-nitrogen (TAN), nitrite-nitrogen, or un-ionized ammonia concentrations in ponds in which bass were stocked at the two densities. These data indicate that largemouth bass of the size used in this study are amenable to pond culture at densities of at least 12,350 fish/ha and that higher stocking densities may be possible.     

POLYCULTURE OF CHANNEL CATFISH (Ictalurus punctatus) WITH ALL-MALE TILAPIA HYBRID

Channel catfish (Ictalurus punctatus) (x? = 0.8 g) and all-male hybrid tilapia fingerlings (Sarotherodon mossambica ♂ × S. hornorum ♀) (x? = 35.0 g) were stocked in 0.04 ha replicated ponds in Baton Rouge, Louisiana, in March and July 1981, respectively. The ponds were stocked at densities of 11,110 catfish per ha, 11,110 catfish with 5,550 tilapia per ha, 7,400 catfish per ha, and 7,400 catfish with 3,700 tilapia per ha. The fish were fed daily at 4% of estimated catfish biomass and were harvested in November 1981. There were no differences in dissolved oxygen or water temperature among the four culture systems (P > 0.05). The presence of tilapia, however, significantly increased water turbidity, pH and chlorophyll a concentrations (P < 0.05). Tilapia did not improve water quality and may have deteriorated it. Tilapia did not affect channel catfish growth or production (P > 0.05), but the presence of tilapia did significantly increase total fish yield (P < 0.05) by 13.5 and 32.2% at low and high catfish densities, respectively. Channel catfish and tilapia averaged 390 and 245 g at harvest, respectively. Overall catfish survival averaged 61%. Tilapia survival was 72% and 61% at low and high densities, respectively.     

Effects of dietary carnitine and lipid on growth and body composition of hybrid striped bass (Morone chrysops ♀× M. saxatilis ♂)

A study was undertaken to investigate the effects of graded dietary levels and different types of carnitine on hybrid striped bass (Morone chrysops × M. saxatilis %) fed different levels of lipid. An incomplete factorial design was utilized in which diets containing lipid at either 5 or 10% were supplemented with l-carnitine at 0, 500, or 1000 mg kg^–1 diet, dl-carnitine at 1000 mg kg^–1 diet, or carnitine chloride to provide 1000 mg carnitine kg^–1 diet. Juvenile hybrid striped bass (3.3 g fish^–1) were stocked into individual 38-l aquaria connected as a brackish water (6), recirculating system and fed each diet in triplicate for 9 weeks.Supplementation of the diet with 1000 mg carnitine kg^–1 increased muscle carnitine from 35.5 to 47.7 g g^–1 tissue. Carnitine supplementation did not result in increased weight gain regardless of carnitine level or type; however, weight gain showed a significant (p<0.05) response to dietary lipid with fish fed diets containing 10% lipid growing 34% more than fish fed diets with 5% lipid. The hepatosomatic index also was unaffected by diet, but the intraperitoneal fat (IPF) ratio was significantly elevated (5.1 vs 3.2%) in fish fed diets with 10% lipid compared to those fed diets with 5% lipid. Fish fed diets containing 1000 mg carnitine kg^–1 had increased IPF ratio values at 4.7% compared to 3.9% for fish fed the basal diet. Liver lipid also was responsive to dietary treatment, increasing from 6.7 to 8.8% of wet weight as dietary lipid increased from 5 to 10%. The relative quantities of triglycerides, free fatty acids and phospholipids in muscle and liver were not influenced by carnitine level, carnitine type or dietary lipid level. Supplementation of carnitine does not appear to be beneficial to hybrid striped bass based on either growth performance or body composition.     

Production of Nile Tilapia Oreochromis niloticus and Freshwater Prawn Macrobrachium rosenbergii Stocked at Different Densities in Polyculture Systems in Brazil

Abstract.— The effect of stocking prawns Macrobrachium rosenbergii at increasing densities in ponds with Nile tilapia Oreochromis niloticus reared at low density was evaluated. Twelve 0.01-ha earthen ponds were stocked with 1 tilapia/m² and 0, 2, 4, or 6 postlarvae prawn/m². Three replicates were randomly assigned to each prawn density. Postlarval prawns were stocked a week prior to tilapia juveniles and both were harvested 175 d after the beginning of the experiment. Tilapia final average weight, survival, production, and food conversion rates did not differ significantly among treatments ( P > 0.05); the averages were 531 g, 67%. 3,673 kg/ha, and 1.91, respectively. Prawn survival rates did not differ for the three stocking densities (mean 90%). However, final weight and production were significantly different ( P < 0.05) as follows: 34.0, 23.0, and 14.7 g and 639, 909, and 818 kg/ha, respectively for 2. 4, and 6 prawns/m² densities. Stocking densities up to 6 prawn/m² did not affect tilapia production and required neither additional feeding nor significant changes in management. The polyculture system allowed an increase in total production with the same amount of supplied feed, thus improving the system sustainability.     

Effects of Stocking Size and Density of Tilapia on Macrobrachium rosenbergii in Polyculture

Studies were conducted to determine the effect of stocking size and density of prawns in polyculture.
In one experiment, postlarval prawns (av. wt. 0.02 g) were stocked in six 0.02 ha earthen ponds at 35,00O/ha. Two ponds were stocked with tilapia fry (av. wt. 0.14 g) and two were stocked with tilapia fingerlings (av. wt. 30.1 g), each at 10,000/ha. Two control ponds had no tilapia. Tilipia stocking size had no effect on prawn growth. Mean weight of prawns after 70 days of culture ranged from a low of 4.5 g when cultured with tilapia fingerlings to a high of 6.6 when cultured in monoculture. Prawn survival was adversely affected by tilapia fry. Average prawn survival in tilapia fry ponds was 65% compared to 75% and 91%, respectively, in tilapia fingerling and monoculture ponds.
In a second experiment, postlarval prawns were stocked in nine 0.02 ha earthen ponds at 40,000/ ha. Six ponds were stocked with 30 g tilapia fingerlings, three at 5,000/ha and three at 15,000/ha. Three control ponds received prawns only. After 100 days of culture, prawn weight ranged from an average of 15.9 g in monoculture ponds to 11.5 g in polyculture ponds. Survival was highest (93.8%) in low density polyculture ponds. Survival was lowest (85.6%) in prawn monoculture ponds. Tilapia reproduction had a negative impact on shrimp production.     

Comparison of stocking time and density on Gulf Killifish growth in freshwater ponds

The Gulf Killifish (Fundulus grandis) is a popular baitfish and a candidate for commercial‐scale production. For inland production, ponds may need to be stocked at different times and densities to meet market demands. Therefore, in Experiment 1, juveniles were stocked into freshwater ponds at a density of 25,000 fish/ha, with two cohorts (mean total length (TL) = 3.92; 3.52 cm) stocked 1 month apart and evaluated for 12 weeks. Cohort 1 had a lower overall survival (40%) than cohort 2 (79%); with no difference between mean final TL (8.46; 8.00 cm); reaching a market size of 6 cm in 22 days and 32 days respectively. In Experiment 2, juveniles (mean TL = 3.16; 3.23 cm) were stocked into freshwater ponds at two densities (25,000 fish/ha and 50,000 fish/ha). Mean final TL was greater in the lower stocking density (8.68 cm) than the higher stocking density (8.20 cm); with no differences in survival between treatments (75; 72%) respectively. Lower stocking and higher stocking densities reached a market size of 6 cm in 29 and 31 days respectively. These results suggest that multiple cohorts can be stocked and grown to market size in one season at a stocking density of 50,000 fish/ha.     

Maximizing First-Year Growth of Mixed-Sex Blue Tilapia Oreochromis aureus in Polyculture with Catfish Ictalurus spp.

Mixed-sex blue tilapia Oreochromis aureus fry were produced from indoor spawns in March and April. Fry were stocked in May at an average weight of 0.5 g at 0, 250, 750, or 2,500/ha into ponds which had been stocked previously with three catfish strains at a combined density of 10,000 fingerlings/ha. After a 167-d culture period, from 15 May until 4 November, tilapia stocked at 250 or 750/ha grew to average harvest size approaching 0.45 kg. No significant difference in average size was observed between males and females except at the 2,500 tilapia/ha stocking density. Sixty percent of the tilapia polyculture ponds yielded no reproduction at harvest and maximum reproduction observed was 13.5 kg/ha. Lack of reproduction and the undeveloped state of sampled female ovaries indicated that most females had yet to attain sexual maturity. Catfish production and catfish feed conversion was not significantly different between tilapia polyculture and catfish monoculture ponds ( P > 0.05). First seine hauls yielded over 88% of the total catfish harvest, but only an average of 3.9% of the marketable tilapia. Tilapia had no measurable impact on the incidence of catfish off-flavor at any of three sample intervals. Overall average levels of TAN and nitrites did not differ significantly between treatments; however, mean chlorophyll a concentration was significantly higher ( P > 0.05) in the 2,500 tilapia/ha treatment than at lower tilapia densities.     

Grass carp (Ctenopharyngodon idella) grazing on duckweed (Spirodela polyrhiza)

The aims of this experiment were (1) toquantify the ability of grass carp to processduckweed and (2) to assess indirect changes inwater chemistry and phytoplankton community,caused by grass carp feeding. Yearling grass carp sized 126 ± 7.7 mm (TL) and19.6 g in weight were kept in 9 laminate tanksof 1 m³ for 14 days. Two stockingdensities (2 and 6 fish per m³) anda control without fish were used. Standard growthrate (SGR) of grass carp fed exclusively onduckweed was 0.70% body weight (BW) d^–1and food conversion ratio (FCR) reached 2.0(average water temperature =21.1 ± 3.8 °C). Daily food intakewas 0.2 g of duckweed dry weight (DW), i.e.,1% of average BW of grass carp. SGR ofduckweed growing in 20 × 20 cm floatingenclosures, differed significantly[F(6,2) = 417.9; p = 0.002] between the twostocking densities of grass carp and thecontrol tanks (without fish). Mean SGR ofduckweed was 0.02 g g^–1 day^–1 and thehighest SGR was recorded in the control tanks.Both decrease in NH₄-N and increase inNO₂-N concentrations differedsignificantly between the treatments[F(2,2) = 45.3; p = 0.02 and F(2,2) = 19.2; p = 0.04 respectively]. Changes in other nitrogenand phosphorus components (NO₃-N, TN, TPand PO₄-P) caused by stocking of grasscarp were not significant. Biomass ofphytoplankton, dominated by filamentous algaeand blue-greens, increased proportionately tostocking density of grass carp. Althoughduckweed has a large potential for nutrientremoval, the most common pathway for thenutrients released through grass carp grazingif duckweed cover is loose is theirincorporation into phytoplankton biomass.     

Growout of hatchery-reared juvenile spotted babylon (<Emphasis Type="Italic">Babylonia areolate</Emphasis> link 1807) to marketable size at four stocking densities in flow-through and recirculating seawater systems

Hatchery reared juvenile spotted babylon, Babylonia areolata, were raised in canvas rearing tanks at four stocking densities (100, 200, 300 and 400 ind. m^–2) in flow-through and recirculating seawater systems until they reached the marketable sizes of 100–140 snails/kg. At all stocking densities, final increments in mean length and weight of snails held in flow-through systems were higher than those in recirculating systems, there was a significant effect (p < 0.05) of culture system on final length and weight. Mean (±SE) survival of snails held at 100 ind. m^–2 in the flow-through system was 100.0 ± 0.1% but was not significantly higher than survival in any other treatment (p > 0.05). Mean survival of snails held in the recirculating system was not significantly lower than at any stocking density in the flow-through system (p > 0.05).     

Strategies for enhancement of natural bay scallop, Argopecten irradians irradians, populations; A case study in the Niantic River estuary, Connecticut, USA

We investigated strategies to enhance populations of bay scallops,Argopecten irradians irradians (Lamarck, 1819), in a presumablyrecruitment-limited natural habitat. At present, the Niantic Riverestuary supports only a minor bay scallop population that is harvestedrecreationally. Three enhancement strategies were evaluated; (1)collection and redistribution of natural spatfall, (2) introduction andover-wintering of hatchery-reared stock into natural habitat to providenew spawning stock, and (3) over-wintering of hatchery-reared stock insuspension culture for creation of mobile spawner sanctuaries. Anassessment of natural bay scallop recruitment in the Niantic Riverconducted in 1997 indicated that few spat were found, they were widelydispersed within the river, and peak spawning occurred in late July1997. Direct re-seeding was evaluated as an enhancement measure byplanting hatchery-reared scallops ( 38 mm shell height) insmall-scale, 100-m² plots at different times and densities.Time of planting and the inferred predation intensity were major factorsaffecting survival; whereas, planting density had no significant effect.Approximately 9,000 scallops (35–45 mm shell height), broadcastwithin an eelgrass bed in November 1997, had high over-winter survivaland underwent gametogenesis and spawning during 1998. Of 26,000 bayscallops ( 45 mm shell height) over-wintered in suspensionculture from 1998–1999, approximately 60–80%survived, and these scallops spawned in mobile sanctuaries, during thesummer of 1999. There is good potential for using aquacultural methodsfor enhancement of bay scallop populations when natural recruitment ispoor and habitat and environmental conditions are not limiting.     

Effects of stocking density on the survival, growth, size variation and condition index of juvenile rabbitfish Siganus rivulatus

Spinefoot rabbitfish, Siganus rivulatus, is an economically important species of herbivorous fish that is relatively easy to rear and thus considered to be suitable for aquaculture. Juveniles are generally reared in nursery systems before being stocked into growout cages or ponds. We report here our evaluation of the effects of stocking density on the survival, growth, feed efficiency and condition index of S. rivulatus juveniles in nursery tanks. The experiment was conducted in a recirculating system of twelve 52-l aquaria connected to a biological filter and a sand filter. Juvenile fish (average weight 6.5 g) were stocked into aquaria at four stocking densities (10, 20, 30, and 40 fish/aquarium) with three replicate aquaria per treatment. Diet was provided at 3% body weight daily divided into two feedings. Fish were weighed weekly for 8 weeks and the diet increased accordingly. Survival was greater than 95% in all treatments, with no significant differences observed among treatments. There were also no differences in specific growth rate (SGR 2.12–2.27) of the fish among treatments. Growth rate was linear during the 8 weeks in all treatments, and harvested biomass increased proportionally to stocking density (198, 401, 600 and 785 g per increasing stocking density, respectively). Feed efficiency (FE 0.67–0.71) of the fish did not vary significantly among treatments. The coefficient of variation was high (35–41%) among the harvested fish, but it also did not differ significantly among treatments. The final condition indices of the fish in all treatments were similar to each other but significantly greater than the initial values (P < 0.05). The results suggest that there is no apparent effect of stocking density at the levels tested on the survival and growth of juvenile rabbitfish.     

Experiments for the production of hybrid striped bass in in-pond circulation systems

Experiments for the production of hybrid striped bass (HSB) in in-pond circulation systems (IPCS) were carried out in 2003 and 2004. The circulation system consisted of two channels with a productive volume of 8.5 cubic meters each. The tanks were installed tightly in a pond, which served for the biological cleaning of the expiry water. In the first year HSB fingerlings with an average weight of 46.4 g were produced. The average yield in the basin was 51.2 kg/m³. The survival rate from stocked 0.44 g advanced fry was 97.8%. The food conversion was 1.16. In 2004 two-year-old HSB were reared in the same IPCS. The tanks were stocked at two different stocking densities, 122 and 244 fingerlings/m³ with a mean weight of 36.5 g. In the tank with the larger stocking density, the yield was almost exactly twice as high as in the other tank (50.0 resp. 24.8 kg), which corresponded to a stocking density of 59.1 or 29.3 kg/m³ at the end of the rearing season. The stocking density had no influence on the increase of the individual body weight. Obviously HSB can therefore still be reared at higher stocking densities.     

Effect of different schedules for broodstock exchange on the seed production of Nile tilapia Oreochromis niloticus (L.) in freshwater

This study compares the effect of three broodstock exchange treatments (exchange every month, two months and three months) versus no-exchange control treatment on year-round seed production and spawning synchrony of Nile tilapia Oreochromis niloticus stocked in 0.4 m³ fiberglass tanks within a recirculating water system. Adult males and females with mean body weights of 161.7 g and 127.6 g, respectively, were stocked at a rate of 4 fish/tank with a male to female sex ratio of 1:3. Water temperature at 29 ± 1.0 °C and photoperiod of 18 h/day were maintained throughout the experiment. The results showed that broodstock in the exchange treatments had higher seed production and spawning synchrony compared with those in the no-exchange treatment. The exchange treatments resulted in 70.5 to 97% increase in seeds kg female^–1 day^–1. The different spawning parameters of broodstock in the two- and three-month exchange treatments were higher than the one-month exchange treatment. Results suggest that broodstock exchange strategy has a long-term improvement effect on seed production and spawning synchrony, and exchanging all breeders every three months is more profitable in terms of better utilization of breeders and in reducing both the holding/conditioning facilities for broodstock and the labor cost.     

Induction of Out-of-Season Spawning of Pikeperch, Sander Lucioperca (L.)

Pikeperch were induced to spawn 3 months prior to the natural spawning period through photothermal and hormonal stimulation. Females (five specimens in each group) were stimulated with injection of human chorionic gonadotropin (hCG) once (200 IU kg^–1), twice (200 IU kg^–1, second dose after 48 h–400 IU kg^–1) or three times (200 IU kg^–1, after 24 h–200 IU kg^–1 and after another 24 h–200 IU kg^–1). The control group was injected once with 0.9% NaCl. The males were stimulated with a single hormone dose of 200 IU kg^–1. Eggs were obtained from all the hormonally treated fish. None of the control group females, which were only stimulated photothermally, ovulated any eggs. The time of ovulation was 66–71 h following the first injection, and the eggs viability until the eyed stage (from 71.5 to 77.5%) did not depend on the number of hormone doses (P > 0.05). The out-of-season spawning method described in this paper could be used to provide pikeperch larvae for intensive culture systems (recirculating water systems) before natural spawning season and to produce larger-sized pikeperch fingerlings for stocking.     

Growth and production of the Amazonian tambaqui in fixed cages under different feeding regimes

Experimental culture of the native Amazonian fish tambaqui, Colossoma macropomum, in fixed cages was carried out over a period of 8 months, in Lake Urubu (Rio Grande do Norte, Brazil), to assess the viability of fixed cage culture of tambaqui and to test the influence of diet on growth rates. Nine synthetic net cages (1 m³) were each stocked with 45-day-old fish (mean weight 3 g; mean total body length 51 mm) at a density of 34 fry m^–3. During the first 2 months of culture, fish were fed a balanced formulated feed on an as-fed basis at the rate of 5% body weight day^–1. During months 3–8 this continued for fish in treatment 1 while those in treatment 2 were fed tropical regional fruits, on a wet weight basis at the rate of 5% body wt day^–1. Fish in treatment 3 were given no supplementary feed. Monthly biometric measurements were made on all fish. Fixed cage fish culture was shown to be a viable and simple technique. Survival in all treatments was 100%. With balanced supplementary feed, production was 14.4 kg m^–3, compared with 4.9 kg m^–3 and 2.1 kg m^–3, respectively, in the treatments where fish were fed with fruits and were not given any supplementary feed.