The need for developing an artificial dry feed for Clarias gariepinus larvae was identified. Dry larval fishfeed requires certain physical properties. Techniques for manufacturing feeds which comply with these requirements were evolved and are recomended for practical use. The techniques that were used for artificial reproduction of C. gariepinus larvae are briefly discussed. The larvae were fed on various dietary formulations and their growth responses were monitored in order to determine their nutritional requirements. A feed was developed which gave superior results to those obtained with natural food organisms. This feed consisted of Torula yeast(69,8%), fishmeal(23,3%), vitamin supplements(O,9%), methionine supplement (150mg/kg), bactericide and preservatives. The feed particles were coated with a lipid supplement which was added at a ratio of 6% to the weight of the feed. The optimum particle size (diameter) for C. gariepinus larvae is 2,2% of their mean total length. The larvae should be given a feed ration of 25% of body weight per day at a frequency of once every four hours. A favourable feed conversion ratio of 1.1g of feed consumed per gram of larval weight gain was experienced. The established dry feed is suitable for use in commercial hatcheries and can be used as a standard and proven food source for further research on the intensive rearing of C. gariepinus larvae.

Inadequate seed production has historically been a factor limiting the commercial culture of Clarias gariepinus. The need to determine the environmental requirements of larvae was identified, in order to facilitate their successful mass rearing in hatcheries. The effects of key environmental factors on the growth, survival and aspects of the behaviour of C. gariepinus larvae and juveniles were investigated under controlled conditions. A strong emphasis was placed on the interpretation of the observed responses in terms of the natural history of the animal. Larvae were obtained by artificially inducing and spawning feral adult broodfish. Larval growth rates were highest in the temperature range 26-33ºC, with a peak at 30ºC. Survival of larvae was high between 22 and 33ºC. The final temperature preferendum of juveniles, determined in a thermal gradient was 30ºC, and it was concluded that this temperature is probably optimal for most of the physiological processes of the animal. Larvae and juveniles displayed strong negative phototaxis. When reared under different photoperiod regimes, larval growth increased with longer dark periods, however a similar rate of mortality was recorded in all photoperiod treatments. The provision of cover was found to enhance larval growth under conditions of continual light (24L/OD), however under conditions of continual darkness (OL/24D) similar growth rates were recorded regardless of the presence or absence of cover. All growth rates in continual darkness (OL/24D) were higher than those in continual light (24L/OD), regardless of whether cover was provided or not. It was concluded that a OL/24D photoperiod is optimal for larval rearing. At salinities between 0 and 5% similar growth and mortality rates of larvae were recorded. At 7.5 % larval growth and survival rates were lower, and at 10 % all larvae died within 48 hours. The 96h-LCSO for unionised ammonia was found to be 2.3 mg/l. The cytological effects of unionised ammonia were evident as degenerative changes in the gill and liver tissues. In an investigation of tank hygiene, it was found that larval growth and survival rates were highest in tanks not cleaned at all. As the frequency of tank cleaning increased, the growth and survival rates of larvae decreased. When reared at densities ranging from 45-450/1, larval growth was found to be density dependent. A model of hatchery productivity (number larvae of standard size produced/time/vol.) was developed based on the density dependent nature of larval growth. The model predicts that maximum productivity, will be achieved at a rearing density of 1400/1. The high growth and survival rates obtained over a broad range of each environmental parameter investigated serve to explain how larvae survive and grow in their unstable floodplain habitat in nature. The broad environmental tolerances of C. gariepinus are of benefit with respect to the culture of larvae, as successful larval rearing is possible over a fairly wide range of environmental conditions.

Based on a number of attributes, the African sharptooth catfish, Clarias gariepinus is an ideal aquaculture species. Much work has been undertaken on developing the technology for the commercial culture of the species inclusive of determining the nutritional requirements for cost effective feed formulation. However, digestible energy (DE) values for specific feed ingredients were unknown. As DE values are not interchangeable between species, this study set out to determine the DE values for conventional and unconventional feed-ingredients for the formulation of an optimal feed. The ultimate goal in feed formulation is to produce a diet that supports the maximum production at the least cost. The indirect method was found to be most suitable in obtaining DE values for the catfish. Feed samples were marked by adding 2% chromic oxide as an inert indicator. The fish were then force-fed and faecal samples were obtained from the hindgut by stripping the fish 10 hours after feeding as this was found to be when the hind-gut was at its fullest and the feed maximally digested. A linear based computer programme was used to formulate experimental diets using the established DE values as an additional variable. The formulated feed was tested and compared to a commercial catfish diet and a standard reference diet (H-440). All of the newly formulated feeds were found to be cheaper to manufacture (by between 5 and 10%). Statistically (ANOVA), all of the experimental diets resulted in higher specific growth rates (SGR) than the standard reference diet. The commercial diet also resulted in a lower SGR than those experimental diets with similar protein levels.

Chiloglanis pretoriae is a rock catlet, indigenous to southern Africa. The aim of the study was to develop a technique to culture C. pretoriae as an alternative to harvesting and selling wild caught fish on the international aquarium trade. As nothing was known about the culture of African rock catlets an investigation into the biology and ecology of the species was necessary to develop the culture protocol. Chiloglanis pretoriae inhabits fast flowing rapids (current speeds over 0.6 metres per second). It is a serial interstitial gravel spawner, that spawns during the summer months. Chiloglanis pretoriae is a carnivorous fish species, feeding on aquatic insects. The natural growth rate is relatively fast in the first two years where after it levels off. Sexual maturity (50%) is attained within the first year (44mm total length). From the four cell stage, embryos took seven days to hatch, 16 days to first feeding and 75 days to reach the juvenile phase. The free embryos were well developed and readily accepted artificial feed at first feeding. The fish spawned readily, without hormone induction, in a continuous raceway. Spawning in the 801 retangular glass aquaria was irregular. The substrate within the raceway consisted of gravel and large rocks. The current was maintained at 0.6m/sec, temperature at 26± 0.6⁰C, dissolved oxygen concentrations at 7.1± 0.3mg/l, pH at 6.9± 0.2 and photoperiod at 16L:8D. Conductivity was monitored and remained within the acceptable range of C. pretoriae (84± 10uS/m).