Aquaculture Research 2015, 46, 346–357 doi:10.1111/are.12181 Utilization of Caridina nilotica (Roux) meal as a protein ingredient in feeds for Nile tilapia (Oreochromis niloticus) James Mugo-Bundi1, Elijah Oyoo-Okoth1,2, Charles C Ngugi3, David Manguya-Lusega1, Joseph Rasowo4, Victoria Chepkirui-Boit1, Mary Opiyo5 & James Njiru1 1 Department of Fisheries and Aquatic Sciences, Moi University, PO Box 1125, Eldoret, Kenya 2 Department of Aquatic Ecology and Ecotoxicology, Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, P.O. Box 9424/1090 GE, Amsterdam, The Netherlands 3 Department of Agricultural Resources, Kenyatta University, P.O. Box 4384, Nairobi, Kenya 4 School of Biological and Physical Sciences, Moi University, P.O. Box 3900, Eldoret, Kenya 5 Kenya Marine and Fisheries Research Institute, Sagana Aquaculture Centre, P.O. Box 451, Sagana, Kenya Correspondence: E Oyoo-Okoth, Department of Fisheries and Aquatic Sciences, Moi University, P.O. Box 1125 Eldoret, Kenya. Email: elijaoyoo2009@yahoo.com Abstract Introduction The effects of replacing fish meal with Caridina nilotica as a protein ingredient on growth performance, nutrient utilization, carcass, proximate composition and economic benefits in Nile tilapia (Oreochromis niloticus) culture was evaluated. Replacement of the FM with C. nilotica was done at 25%, 50%, 75% and 100% (D25, D50, D75 and D100) and the substitution effects was compared with the control diet (D0, 0% C. nilotica). After 140 days of culture, the best growth performance, nutrient utilization and economic benefits occurred in fish groups fed diets with 25% C. nilotica inclusion. However, growth performance in fish fed diets D50 and D75 were comparable with the control (P > 0.05). At 100% substitution level of FM with C. nilotica, the growth performance and fish survival was lower than control. Protein and lipid contents in the fish and their digestibilities were highest in diet D25 and decreased with increasing levels of substitution of FM with C. nilotica. This study demonstrate that utilization of local protein sources (C. nilotica) can be effectively used to replace up to 75% of FM in the diets without compromising growth performance, survival, nutrient utilization and economic benefits in O. niloticus culture. Protein remains the most expensive ingredient in prepared feeds for most cultured organism, yet it is also the most important factor affecting growth performance of fish. Therefore, one of the foreseen constraints to intensification of fish farming is the scarcity of inexpensive, readily available and nutritive sources of protein. Fish meal (FM) is a major protein source in aquafeeds for most species of fish, because it is an excellent source of essential nutrients such as indispensable amino acids, essential fatty acids, vitamins, minerals, attractants and unknown growth factors (Tacon 1993; Abdelghany 2003). However, the pitfalls of continued utilization of FM in aquafeed are now widely recognized; FM is in limited supply and more expensive than most other protein sources (Tacon & Metian 2008). The aquaculture industry use 40% of the global fish meal, yet fish meal production has remained stable from the late 1980s at about 6 million metric tonnes/annum (FAO 2006). Moreover, FM usage in aquaculture for 1999 was over 2 million metric tonnes and is estimated to reach well over 4 million metric tonnes by 2015 (New & Wijkstro€ om 2002). The consensus is that alternative protein sources are needed to supplement or replace FM in aquafeeds, thus contributing to long-term sustainability of the aquaculture industry (see Tacon & Jackson 1985; and reviews Keywords: Oreochromis niloticus, Caridina nilotica, Growth, FCR, Nutrient utilization 346 © 2013 John Wiley & Sons Ltd Aquaculture Research, 2015, 46, 346–357 in Gatlin, Barrows, Brown, Dabrowski, Gibson, Gaylord, Gaylord, Hardy, Herman, Hu, Krogdahl, Nelson, Overturf, Rust, Sealey, Stoneberg & Souza 2007). Several studies have evaluated plant and animal protein sources as possible FM replacement or supplements (Gaber 1996; Pouomogne, Takam & Pouemegne 1997; El-Sayed 1998; Fagbenro 1998; Fontainhas-Fernandes, Gomes, Reis-Henriques & Coimbra 1999; Maina, Beames, Higgs, Mbugua, Iwama & Kisia 2002; Abdelghany 2003; El-Saidy & Gaber 2003; Richter, Siddhuraju & Becker 2003; Hern andez, Olvera-Novoa, Hardy, Hermosillo, Reyes & Gonz alez 2010; OliveraCastillo, Pino-Aguilr, Lara-Flores, Granados-Purto, Montero-Munoz, Olvera-Novoa & Grant 2011; Richie & William 2011). The results show great variation in the degree of success for partial or complete substitution depending on the ingredient of the test feeds as well as species of fish under culture. Caridina nilotica (Roux) is currently an important prey item for many fish species in Lake Victoria, the second largest fresh water lake in the world (Budeba & Cowx 2007). The anaerobic environment in L. Victoria has favoured the flourishing of C. nilotica and other micro-invertebrates. It was observed in trawl catches, in the stomachs of Nile perch and is an important by-catch in the Rastrineobola argentea (dagaa) fishery by light attraction in Lake Victoria (Balirwa, Chapman, Chapman, Cowx, Geheb, Kaufman, Lowe-McConnell, Seehausen, Wanink, Welcomme & Witte 2003; Matsuishi, Muhoozi, Mkumbo, Budeba, Njiru, Asila, Othina & Cowx 2006). Since 1986, the abundance of C. nilotica in the waters of Lake Victoria has increased tremendously, although few quantitative records are available (Cowx, Van der Knaap, Muhoozi & Othina 2003; Matsuishi et al. 2006). The average Caridina biomass for the whole lake was estimated at 22 694 metric tonnes by hydroacoustic surveys (Getabu, Tumwebaze & MacLennan 2003). In Kenya, the livestock feed industry recognized the underutilized status of C. nilotica and have incorporated it as a dietary protein on subsistence scale (Munguti, Waidbacher, Liti, Straif & Zollitsch 2009). It is also being used as bait in haplochromine hand-line fisheries, albeit the haplochromine fishery has declined tremendously over the years (Budeba 2007). Previous attempts at using the ingredient in aquaculture were highly promising for both the adult Nile tilapia (Oreochromis niloticus) © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Utilization of Caridina nilotica Mugo-Bundi et al. (Liti, Waidbacher, Straif, Mbaluka, Munguti & Kyenze 2006) and larval stages of the African catfish (Clarias gariepinus) (Rasowo, Ngugi & Macharia 2008; Chepkirui-Boit, Ngugi, Bowman, OyooOkoth, Rasowo, Mugo-Bundi & Cherop 2011). This by-catch as yet is underutilized and can be profitably exploited as a source of protein in the aquaculture industry. This study was undertaken to determine the effects of incorporation of C. nilotica as a replacement of FM on the growth performance, survival, feed and nutrient utilization as well as digestibility in O. niloticus culture. Oreochromis niloticus is a typical omnivorous warm water fish species, and the production of tilapia had been over 2500 thousand tonnes per year in the world (FAO 2006). The species also feed on variety of food items (Pullin 1996) thus offering a possibility for testing the suitability of C. nilotica as a protein ingredient in aquafeed. We used the C. nilotica to replace FM to further the development of aquaculture using locally underutilized feed ingredients. Materials and methods Fish and experimental setup The experiment was carried out under controlled hatchery conditions at Moi University, Eldoret Kenya in the Department of Aquaculture and Fisheries Science. Three mature female broodstock (mean weight = 398  7.5 g) and two mature males (mean weight = 460  8.1 g) were selected based on the method of Viveen, Richter, van Oordt, Janssen and Huisman (1985) and transferred to the hatchery. Larvae were obtained through induced breeding and semi-natural spawning. During the culture period, the larvae were fed Artemia nauplii. The larvae were cultured for a period of 21 days to an initial mean weight of 24.0  2.0 g in a flow-through raceway-type 2500-l open water tanks, supplied with filtered dechlorinated tap water at a rate of approximately 50 L h 1. The water was continuously aerated, and temperature was controlled thermostatically at 27.0  0.5°C. Feeding was carried out using a commercial extruded tilapia feed (Raanan Fish Feed Co., Israel: crude protein 270 g kg 1; crude lipid 56 g kg 1; crude fibre 61 g kg 1; ash 62 g kg 1; NFE, 551 g kg 1). After acclimation, all the fish were mixed in one tank and randomly distributed between eighteen 347 Utilization of Caridina nilotica Mugo-Bundi et al. 500-L cylindroconical tanks supplied with filtered, dechlorinated tap water (salinity determined by salinometer (Model IC/SB-1 Salinity Cell) was 0.3 psu; NO2 < 0.06 mg L 1; NO3 <0.01 mg L 1; NH3 < 0.02 mg L 1; pH 7.2). Testing conditions included 400 fish per tank, with each formulated diet being experimentally tested in triplicate. Fish were held under natural light, with a photoperiod regime of 12-h light and 12-h dark (0°34′13.8″N and 35°18′49.8″E) at a constant temperature 27.1  0.2°C maintained using thermostat heaters. The flow-rate was constantly regulated at about 20 L h 1 to maintain dissolved oxygen above 80% of the saturation level. The fish were then cultured for 140 days. Chemical analyses Dietary ingredients, formulated diets, faeces and whole bodies of fish samples were analysed for proximate composition according to the standard methods of AOAC (1990). The ingredients were analysed to determine the exact quantity needed for the formulation of the diets. Moisture content was estimated by drying the samples to constant weight at 105°C in a drying oven (GCA, model 18EM, Precision Scientific group, Chicago, IL, USA) and nitrogen content using a microKjeldahl apparatus (Labconco Corporation, Kansas, MO, USA). Crude protein was estimated as N 9 6.25. Lipid content was determined by ether extraction in a multi-unit Soxhlet extraction apparatus (Lab-Line Instruments, Inc., Melrose Park, IA, USA) for 16 h. Ash was determined by combusting dry samples in a muffle furnace (Thermolyne Corporation, Dubuque, Iowa, USA) at 550°C for 6 h. Crude fibre were analysed by Weende methods. Nitrogen Free extracts (NFE) were determined by the difference method. For amino acid determination, samples were hydrolyzed with 6 M HCl at 110°C for 24 h. Sulphur-containing amino acids (cysteine and methionine) were oxidized using performic acid before acid hydrolysis. Amino acids were separated using reverse phase, HPLC and quantified following post-column derivatization within ninhydrin. All analyses were performed in triplicates. Gross energy of the diets and faeces was determined using an adiabatic bomb calorimeter 1241, Parr Instrument Company, MolineIllinois-USA). The analysed composition of the fish meal and C. nilotica used in the feed formulation are presented in Table 1. 348 Aquaculture Research, 2015, 46, 346–357 Table 1 Analysed chemical composition and essential amino acid of the fish meal and Caridina nilotica used in formulating the experimental diets Composition (g kg 1 as fed) Fish meal (Rastrineobola argentea)* DM† Crude protein Crude lipid Ash Crude fibre NFE‡ Gross energy (kJ g 1)§ 894 671 95 61 25 42 19.8 Essential amino acid profiles (g 100 g 1 diet) Arginine 6.01 Histidine 1.7 Isoleucine 4.01 Leucine 6.52 Lysine 5.45 Methionine + Cystine 4.59 Phenylalanine + Tyrosine 6.73 Threonine 3.53 Tryptophan 1.82 Valine 4.07 Caridina nilotica* 914 561 105 98 75 75 18.2 4.42 1.41 3.61 5.71 3.51 2.45 3.79 3.41 1.52 4.08 *Obtained locally †DM, dry matter ‡Nitrogen free extract = 1000 (moisture content + crude protein + crude lipid + ash + fibre). §GE (gross energy): calculated using conversion factors 23.0, 38.1 and 17.2 kJ g 1 for protein, lipids and carbohydrates (Tacon 1990). Feed formulation and feeding regimes Five isonitrogenous and isocaloric diets (crude protein 27%, Gross Energy 17.7 kJ g 1) were formulated. The C. nilotica was used as protein ingredient to replace the FM protein at 25%, 50%, 75% or 100% (designated as D25, D50, D75 or D100, respectively) control diet (D0) was formulated with FM as the sole protein ingredient. The C. nilotica was oven-dried at 30°C for 12 h before being ground using an electric meat grinder (Model: SM-G70; Guangzhou Sunmile Industries, China). The ingredients proportions and proximate compositions of the experimental diets are shown in Table 2. A reference tilapia diet (DRF) purchased from Raanan Industries Israel described above was used to compare the fish performance with that of our experimental diets. The experimental diets were prepared by cooking extrusion using semiindustrial extruder (Hobart M-600; Hobart Corp., Troy, OH, USA). Perch oil, mineral and vitamin premixes was gradually added and a warm water © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Aquaculture Research, 2015, 46, 346–357 Utilization of Caridina nilotica Mugo-Bundi et al. Table 2 Ingredients and chemical composition (g kg of experimental diets used for feeding O. niloticus 1 ) Diets D0 D25 D50 D75 D100 320.0 0.0 297.0 296.0 240.0 98.0 295.0 279.0 160.0 196.0 301.0 254.0 80.0 294.0 317.0 218.0 0.0 392.0 298.0 216.0 10.0 20.0 20.0 20.0 12.0 5.0 11.0 20.0 20.0 20.0 12.0 5.0 12.0 20.0 20.0 20.0 12.0 5.0 14.0 20.0 20.0 20.0 12.0 5.0 17.0 20.0 20.0 20.0 12.0 5.0 928.2 276.3 52.3 68.0 59.0 544.4 17.7 919.1 276.3 54.3 71.0 62.0 536.4 17.7 917.5 276.7 56.4 74.0 66.0 526.9 17.6 921.5 276.2 58.8 77.0 70.0 518.0 17.5 1 Ingredients (g kg diet) Sardine fish meal Caridina nilotica Wheat floor Corn grain (extrusion-cooked) Perch liver oil Binders (Cassava) Vitamin premix* Mineral permix† Salt (Nacl) Chromic oxide (Cr2O3)‡ Chemical analysis (g kg Dry matter Crude protein Crude lipid Ash Crude fibre NFE§ Gross energy (kJ g 1) 1 DM) 923.0 276.6 50.1 63.0 58.0 552.3 17.8 *Commercial formula (mg premix kg 1 diet). Vitamins (mg): retinol, 1000; thiamine, 1200; riboflavin, 2000; pyridoxine, 1000; cyanocobalamine, 200; ascorbic acid (Stay C), 5000; cholecalciferol, 2400; a tocopherol, 1000; pantothenic acid, 400; choline chloride, 1600; folic acid, 2500; nicotinic acid, 1800; biotin, 1200; inositol, 3000; paraminobenzoic acid, 3200. †Minerals (mg): cobalt, 400; copper, 2100; iron, 2000; iodine, 1600; manganese, 4000; zinc, 2000; selenium, 400. ‡ICN Corporation, Costa Mesa CA. §NFE (nitrogen free extracts) = 100 (protein% + lipid % + ash% + fibre%). (approximately 50% of the total weight) was added to facilitate mixing and homogenization. The diet was then prepared into a paste, which was separately passed through a grinder, and cold-pelleted (1-mm diameter) using Simon-Heese pelleting machine (Boxtel, the Netherlands). They were dried in a forced-air drier at room temperature for 24 h and stored in plastic bags at 4°C for further use. Table 3 contains the calculated AA profiles of the experimental diets according to NRC (National Research Council) (1993). Each of the five experimental diets together with the reference diet were randomly assigned to three groups (i.e. triplicate tanks) of tilapia. The fish were fed at 4% of the biomass of each tank divided into two feedings (0800 and 1700 h) 7 days a week. Any uneaten feed was then collected from the tank after the feeding experiment. The rations were adjusted according to the amount of unconsumed feed. Water quality measurements Water samples were collected fortnightly from each tank. Dissolved oxygen (DO), temperature and pH were measured in the tanks using a calibrated JENWAY 3405 electrochemical analyser (Barloword Scientific Ltd, Essex, United Kingdom). Unionized ammonia was measured using DREL/2 HACH kits (HACH Co., Loveland, CO, USA). In all treatments, dissolved oxygen concentrations ranged from 6.9 to 7.2 mg/L, unionized ammonia <0.02 mg L 1 and pH ranged from 7.3 to 7.8. All the water quality parameters were within the acceptable ranges for fish growth (Boyd 1984). Table 3 Calculated essential amino acid (EAA) composition of the test diets used (g 100 g 1 diet) Diets Amino acid DRF D0 D25 D50 D75 D100 *NRC (National Research Council) (1993): requirement Arginine Histidine Isoleucine Leucine Lysine Methionine + Cystine Phenylalanine + Tyrosine Threonine Tryptophan Valine 1.65 0.64 1.08 2.02 1.64 0.99 1.66 0.88 0.46 1.35 1.64 0.63 1.06 2.03 1.65 0.96 1.65 0.87 0.43 1.34 1.56 0.62 1.04 2.01 1.64 0.91 1.63 0.85 0.44 1.26 1.44 0.54 0.97 1.95 1.55 0.76 1.52 0.84 0.42 1.24 1.31 0.51 0.81 1.92 1.47 0.68 1.43 0.82 0.45 1.22 1.10 0.52 0.76 1.87 1.41 0.62 1.40 0.81 0.42 1.21 1.20 0.42 0.73 0.98 1.43 0.64 1.40 0.56 0.14 0.84 *Amino acid requirement according to NRC (National Research Council) (1993). DRF Reference diet. © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 349 Utilization of Caridina nilotica Mugo-Bundi et al. Aquaculture Research, 2015, 46, 346–357 Sampling and evaluation of growth parameters Statistical analyses A total of 20 fish, were sampled from each tank using dip nets and bulk weighed in a batch of five fish, every 2 weeks to calculate weight gain and feed conversion ratio. Body composition was determined by sampling three fish from each replicate tank, at the beginning and after the 140 days feeding trial. Once measurements had been taken, all samples were frozen at 4°C until analysed. The effects of diets on growth were determined by evaluating a number of growth and nutrient utilization indices, including weight gain, specific growth rate (SGR), survival, feed conversion ratio (FCR), protein efficiency ratio (PER), protein productive value (PPV) and protein growth rate (PGR). The following formulas were used: Weight gain ? Final mean fish weight Initial mean fish weight SGR ? (e.g 1) 9 100, where g = (ln(W2) (ln(W1))(t2 t1) 1 and W2 and W1 are weights on day t2 and t1, respectively. FCR ? Feed intake/weight gain PER weight gain (g)/protein intake (g); PPV (%) = 100 9 (protein gain (g)/protein intake (g)); PGR%/day) = 100 9 (Ln final protein content Ln initial protein content)/days of feeding. The digestibility study was carried out during the last month of the experiment. For digestibility tests, 0.5% chromic oxide was included in the experimental diets. After seven-adaptation period, faeces were collected using a modified faecal collection system for 28 days, 7 days a week, centrifuged (4°C, 4000 rpm, 15 min), freeze-dried and used to analyse the natural marker AIA (Acid insoluble ash). Apparent digestibility coefficients (ADC) were calculated using the formula as follows: Statistical analyses were performed using SPSS version 17.0. The effect of experimental diets on growth, survival, FCR, nutrient utilization and carcass composition were performed by analysis of variance (One-way ANOVA). When significant differences were discerned, treatment means were compared using Duncans Multiple Range Test (DMRT). Values throughout the text are expressed as mean  standard error. In all the analysis, significant was accepted at P < 0.05. An enterprise was used to determine the revenue, costs and returns of the feeds. The profitability of the enterprise was analysed using the net returns above variable costs. The break-even price was calculated using the formula    % dietary Cr2 O3 ADCnutrient ð%Þ ¼ 100  1 % faecal Cr2 O3   % faecal nutrients  % dietary nutrient ADC of gross energy was calculated using gross energy data (kJ g 1) instead of per cent nutrient data.  ADCdrymatter ð%Þ ¼ 100  1 350   % dietary Cr2 O3 % faecal Cr2 O3 Breakeven price ¼ Fixed cost per unit 1 ðVariable cost per unit=Selling Price per unitÞ Results Growth performance and survival The overall values of growth performance (in terms of final mean weight, weight gain and SGR) are shown in Table 4. Parameters of growth performance were affected by substitution levels of C. nilotica during the grow-out period. Highest final mean weight, weight gain and SGR were obtained in fish fed diet D25. Parameters of growth performance in diets D50 and D75 were statistically similar to those of the control group and to the reference diet (P > 0.05). However, at 100% substitution of FM with C. nilotica, all values of growth performance decreased. Survival was significantly (P < 0.05) different among treatments with highest value recorded in all the diets except D100. Trend curves for growth of O. niloticus under different feed treatments for the entire growth period are shown in Fig. 1. The growth trends in fish fed reference diet (DRF) were similar to the growth trends in fish fed D0 (data not shown). Trend in growth of the fish fed diet D25 were higher than control diet (D0), D50 and D75 (but not statistically) and maintained similar trends in growth, but which were significantly (P < 0.05) higher than growth trends in fish fed experimental diet D100. © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Aquaculture Research, 2015, 46, 346–357 Utilization of Caridina nilotica Mugo-Bundi et al. Table 4 Growth performance of O. niloticus under different diets formulated using C. nilotica Diets Growth performance parameters D0 DRF D25 D50 D75 D100 24.4  1.3 24.9  0.9 24.9  1.0 24.6  1.2 24.8  1.3 24.4  1.1 Final mean weight (g) 544.2  20.4b 540.8  26.5b 573.0  15.2c 519.1  33.1b 507.0  33.8b 392.0  53.7a Weight gain (g) 519.8  20.0b 515.6  23.4b 548.1  14.7c 495.1  26.9b 482.4  33.2b 367.2  49.2a 2130.2  122.1b 2.25  0.07b 2113.4  119.2b 2.21  0.08b 2201.2  123.6c 2.24  0.06c 2071.5  126.2b 2.20  0.08b 1960.9  142.4b 2.16  0.06b 1480.7  152.6a 1.97  0.08a 93.4  2.2b 93.2  2.1b 93.3  1.8b 96.5  2.8b 95.0  1.9b 81.0  5.3a Initial mean fish weight (g) % weight gain Specific growth rate (SGR;% day 1) % survival Means with the same letters as superscripts are not significantly different (P > 0.05). Values are expressed as mean  standard error. SE: standard error, calculated from the mean-square for error of the ANOVA. Mean body weight (g) 600 500 D25 D0 D50 D75 400 D100 300 200 100 0 0 20 40 60 80 100 120 140 Time (days) Figure 1 Growth curves for Oreochromis niloticus in various feed treatments during this study period. D0 is control diet (formulated with 100% FM as the sole protein) whereas D25, D50, D75 and D100 represent substitution of FM with 25%, 50%, 75% and 100% Caridina nilotica respectively. Vertical bars denote  standard error of the mean. Nutrient utilization Parameters of nutrient utilization in fish under experimental diets are shown in Table 5. Again parameters of nutrient utilization in diet D0 and DRF were similar (P > 0.05). Although the overall FCR was lowest in fish fed D25 and tended to increase with increased substitution levels of FM with C. nilotica, there were no significant differences in FCR in fish fed diets D0, D25, D50 and D75. The PER, PPV and PGR were highest in diet D25, however, fish fed diets D50 and D75 as well as those fed control diets had similar nutrient utilization parameters (P > 0.05). Whole carcass composition Data on whole body composition are presented in Table 6. All the values of carcass proximate © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 composition between the reference diet and D0 were similar (P > 0.05). No significant changes in moisture content were observed at the different dietary treatments (P > 0.05). However, the protein contents in whole-body fish was significantly (P < 0.05) highest in diet D25, but decreased thereafter with increasing levels of C. nilotica inclusion. Whole body lipid contents were comparable at 0, 25, 50 and 75% levels of substitution, but at 100% substitution levels, the lipid content in fish was significantly (P < 0.05) the lowest among the dietary treatment. The fibre content in the fish was comparable in all dietary treatments except at 100% inclusion levels of C. nilotica. Ash content in the fish increased significantly (P < 0.05) with increased inclusion levels of C. nilotica in the formulated diets. Nutrient digestibility Apparent nutrient digestibility was high for protein, lipid and energy and showed significant variation among the dietary groups (P < 0.05) (Table 7). Protein digestibility was highest for diet D25 and D50, which were not significantly different (P > 0.05) from the diet D0. Lipid digestibility was comparable for all the diets except D100, which was found to be low. Digestible carbohydrates and dry matter were similar for all diets (P > 0.05). Economic analysis Yield of fish under diets containing different inclusion levels of C. nilotica and the enterprise budget for different treatments is provided in Table 8. Highest fish yield was obtained when feeding was 351 Utilization of Caridina nilotica Mugo-Bundi et al. Aquaculture Research, 2015, 46, 346–357 Table 5 Parameters of feed and nutrient utilization of O. niloticus under different diets formulated using C. nilotica Diets Nutrient utilization parameters D0 DRF 1 Feed intake (g feed g fish) Feed conversion ratio (FCR) Protein efficiency ratio (PER) Productive protein values (PPV;%)) Protein growth rate (PGR;% day 1) 641.4 1.25 2.97 26.2     33.2 0.35a 0.46c 1.12b D25 649.4 1.25 2.94 25.2 0.30  0.03b     31.2 0.39a 0.59b,c 5.31b 653.7 1.19 3.11 28.4 0.29  0.02b D50     26.5 0.33a 0.25c 3.33c D75 651.1 1.28 2.89 24.8 0.32  0.03c     27.3 0.31a 0.21b 4.33b D100 666.8 1.31 2.82 24.3 0.27  0.04b     30.2 0.31a 0.24b 4.43b 0.29  0.03b 649.7 2.03 1.82 15.47     31.9 0.41b 0.15a 3.25a 0.18  0.02a Means with the same letters as superscripts are not significantly different (P > 0.05). SE: Standard Error, calculated from the mean-square for error of the ANOVA. Table 6 Carcass proximate composition of O. niloticus fed experimental diets Diets Chemical analysis Initial value DRF Moisture Protein Lipids Fibreere Ash 77.1 10.2 4.2 4.2 3.1      76.9 15.3 5.4 3.8 3.3 5.6 1.1 0.9 0.7 0.5 D0      6.7 0.2b 0.3b 0.3a 0.2a 75.9 15.4 5.3 4.0 3.4 D25      6.7 0.2b 0.3b 0.3a 0.2a 75.7 15.7 5.4 4.1 3.3 D50      8.1 0.3c 0.2b 0.6a 0.5a 74.3 14.9 5.3 4.3 3.8 D75      5.7 0.3b 0.4b 0.2a 0.4b 75.2 14.6 5.1 4.5 4.0 D100      5.5 0.3b 0.3b 0.2a 0.4b 75.5 13.2 4.8 5.6 4.5      8.8 0.4a 0.4a 0.5b 0.4c Values with the same letters as superscripts in the same row are not significantly different (P > 0.05). Comparisons were made between dietary treatments and excluded the initial values. Table 7 Apparent digestibility coefficients (ADCs) of experimental diet components Diets Apparent digestibility DRF Dry matter Protein Lipids Energy 72.4 95.2 91.3 83.1 D0     10.1 3.4c 7.4b 5.2 73.4 94.2 90.4 81.2 D25     11.2 3.3c 7.7b 5.6 74.8 95.5 92.1 80.9 D50     9.9 4.7c 8.5b 7.6 75.4 94.3 94.3 81.7 D75     8.9 4.2c 3.7b 5.9 72.8 90.1 92.1 80.5 D100     7.8 3.2b 4.9b 5.8 73.9 83.2 84.2 78.5     10.2 5.1a 5.2a 6.4 Values with the same letters as superscripts in the same row are not significantly different (P > 0.05). performed using diet D25 and thereafter increased inclusion of C. nilotica in the diet resulted in reduced yields. The lowest total fish yields occurred in treatments with 100% C. nilotica. The total investment and operational costs were highest at treatment D25 and thereafter reduced with increasing inclusion of C. nilotica in our formulated diet. Net returns above both the total cost (TC) and total variable cost (TVC) were significantly better in fish reared using diet D25. Increasing inclusion of C. nilotica beyond 25%, 352 resulted in decreased net returns above TVC and TC decreased, but the enterprise was still profitable until 75% C. nilotica inclusion levels. There were negative net returns above TVC and TC at when feeding was carried out using test diet D100. With exception of diet D100, the break-even price in all diets were below the selling price of fish locally (US $ 2.1) with treatment D25 registering the lowest break-even price. However, for diet D100, the investment posted negative returns at a selling price of US $ 2.1. © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Aquaculture Research, 2015, 46, 346–357 Utilization of Caridina nilotica Mugo-Bundi et al. Table 8 Enterprise budget (in US $) of O. niloticus under different diets formulated using C. nilotica Parameters Tank capacity (L) Final weight of fish Survival Harvest weight Unit cost Gross receipts Variable costs Tilapia fingerling costs Cost of feeds Field labour Cost of equipment Electricity Miscellaneous Sub-total variable costs Interest on operating cost Total variable cost (TVC) Fixed costs Tank costs Amortization Interest on fixed cost Total fixed cost Total cost (TC) Net returns above TVC Net returns above TC Break-even price DRF D0 D25 D50 D75 D100 500.0 544 93.4 12,199 2.1 25,617 500.0 541 93.2 12,097 2.1 25,403 500.0 573 93.3 12,831 2.1 26,944 500.0 519 96.5 12,022 2.1 25,247 500.0 507 95.0 11,560 2.1 24,275 500.0 392 81.0 7,620 2.1 16,003 5,040 4,628 1,800 2,400 960 1,200 16,028 2,564 18,592 5,040 4,597 1,800 2,400 960 1,200 15,997 2,559 18,556 5,040 4,813 1,800 2,400 960 1,200 16,213 2,594 18,807 5,040 4,455 1,800 2,400 960 1,200 15,855 2,537 18,392 5,040 4,382 1,800 2,400 960 1,200 15,782 2,525 18,307 5,040 3,605 1,800 2,400 960 1,200 15,005 2,401 17,406 2560 1000 180 3,740 22,332 7,025 3,285 1.12 2560 1000 180 3,740 22,296 6,847 3,107 1.15 2560 1000 180 3,740 22,547 8,137 4,397 0.97 2560 1000 180 3,740 22,132 6,855 3,115 1.15 2560 1000 180 3,740 22,047 5,968 2,228 1.32 2560 1000 180 3,740 21,146 1,403 5,143 5.60 Major budjet assumptions: Interest rates on fixed cost = 18% Commercial production will be conducted in 60 tanks Production assumption: Stock advanced fingerlings to grow-out in one season per year. Land assumption: Own the land, no other economic use, only land charge for property taxes. Discussion In this study, the crude protein (29.7%) and energy (17.7 kJ g 1) of the experimental diets were formulated based on the protein and energy requirements of Nile tilapia as suggested by AbdelTawwab, Mohammad, Ahmad, Khattab and Shalaby (2010). We compared our experimental control diet with a reference commercially available diet containing the same protein and energy content as our diets. The results of this study demonstrates that up to 75% of fish meal can be replaced with C. nilotica in a formulated diet without negative effects on growth performance, survival, feed and nutrient utilization during the culture of O. niloticus. The growth performance of O. niloticus in terms of final mean weight at harvest, weight gain and SGR under varying inclusion levels in this study is comparable with other studies (Gaber 1996; El-Sayed 1998; El-Saidy & Gaber 2003; AbdelTawwab et al. 2010; Hernandez et al. 2010). In © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 these studies, average per cent weight gain ranged from 297% to 2634% and SGR ranged from 1.4% to 5.1% day 1. The observed growth response of the Nile tilapia presumably reflects the high digestive capacity of this species for a wide range of food items (Degani, Viola & Yehuda 1997). There are few studies available evaluating the growth performance of Nile tilapia when fed diets containing C. nilotica (e.g. Liti et al. 2006). Our data showing O. niloticus attaining up to 480-540 g when cultured for 140-day period demonstrate that diets formulated using C. nilotica does not compromise the overall growth of O. niloticus. In a previous experiment, using the C. niloticus protein source, we had demonstrated the growth performance of juvenile C. gariepinus fed diets formulated with C. nilotica was comparable with fish fed Artemia salina (Chepkirui-Boit et al. 2011). In this study, diets formulated using C. nilotica as the main protein source resulted in the best growth performance at 25% inclusion level, whereas at 50% and 75% inclusion of C. nilotica, the growth 353 Utilization of Caridina nilotica Mugo-Bundi et al. performance were still comparable with the control (and the reference) diets. Also, the trend in growth of O. niloticus fed diets D0 (and reference diet DRF), D25, D50 and D75 were similar during entire growth period. These observed trends were attributed to the combined nutritive values found in FM and C. nilotica and the ability of O. niloticus to utilize the available nutrients. Indeed, it has been reported that the growth of fish fed diets containing mixed rations, depends on the nutrient composition of the individual feed components and the ability of the animal to digest and absorb the combined nutrients (Degani et al. 1997; Falaye & Jauncey 1999; Riche, Trottier, Ku & Garling 2001). The reduction in growth performance and survival at 100% substitution levels and lower growth trend curves for diet D100 may probably indicate limitation of essential amino acid such as arginine, lysine and methionine + cysteine (Table 2). However, more research is necessary to support this hypothesis. Therefore, 100% substitution of FM with C. nilotica appears impractical. Results of this study further demonstrate that the parameters of nutrient utilization were affected by inclusion levels of C. nilotica. The current FCR values coincided with previously published ranges for Nile tilapia (Al-Hafedh 1999; Abdelghany 2000; Khattab, Ahmad, Shalaby & Abdel-Tawwab 2000; Abdel-Tawwab et al. 2010). The feed conversion ratio (1.2–1.4), protein efficiency ratio (2.8–3.2) values recorded here at C. nilotica inclusion levels of up to 75% are better than those reported by El-Sayed (1998) (FCR = 1.86–2.24, PER = 1.55–1.70) in a study of other animal protein sources such as poultry by-product meal (PBM) and meat and bone meal (MBM) in Nile tilapia diets. Our data were comparable to those of Fasakin, Serwata and Davies (2005) of hybrid tilapia O. niloticus 9 O. mossambicus fed diets with partial replacement of FM with PBM. The best PER, PPV and PGR were all obtained at an inclusion level of up to 25% C. nilotica in the diet, which point to a higher protein intake efficiency because of combination of proteins sources from two ingredients. With exception of fish fed at 100% C. nilotica inclusion, PER values in all treatments were higher than 2, indicating efficient protein utilization. However, at inclusion levels of up to 75% the parameters of nutrient utilization were comparable with control diets (P > 0.05) and therefore points to efficient nutrient utilization parameters even at high levels of FM substitution. 354 Aquaculture Research, 2015, 46, 346–357 In this study, the carcass proximate analysis of fish (except moisture content) was significantly influenced by FM substitution levels. Protein and lipid contents in O. niloticus were highest for diet D25 and reduced with increasing C. nilotica inclusion level. The observed changes in protein and lipid contents in fish body could be linked with changes in their synthesis, deposition rate in muscle and/or different growth rates (Soivio, Niemisto & Backstrom 1989; Abdel-Tawwab, Khattab, Ahmad & Shalaby 2006). Because of the lower protein content of the replacement diets (compared to FM), the reduced digestibility of the diets containing high level substitution of C. nilotica, the high ash content and generally higher fibre content in the diets containing high level C. nilotica could possibly affect protein conversion by the fish. However, a reduced physiological ability of the fish to convert the protein and lipids in the food into body proteins and fats ,respectively, is also plausible. The ADCs for all the experimental diets except D100 were high (Table 7). To the best of our knowledge, there is no study on the digestibility of diets formulated using C. nilotica. However, the current values are comparable with those reported for herring meal (94.9%), menhaden meal (89.9%) and PBM (95.9%) for rainbow trout (Oncorhynchus mykiss) diets (Sugiura, Dong, Rathbone & Hardy 1998). They were higher than values for crude protein ADC for poultry offal meal (74%) (Hanley 1987). The high protein digestibility values found in thisstudy reflect high-quality raw materials used to formulate the feeds together with the excellent amino acid profiles of the test ingredient used as protein source. This is because the nutritional value of a protein feedstuff is primarily based on its essential amino acid content and bioavailability (Dias, Alvarez, Diez, Arzel, Corraze, Bautista & Kaushik 2005). In addition, the high ADC crude protein values registered confirm Nile tilapia’s ability to digest C. nilotica protein. Highest total fish yield was achieved in treatments with diet D25 and with increased inclusion of C. nilotica in the diet, yields of O. niloticus reduced. At this dietary inclusion level, highest yield were accounted for by the better yields obtained from the fish growth. Higher nutrient digestibility of the diets could explain the increased yields at this inclusion level as already noted. In this study, the total investment and operational costs were affected by dietary treatments. For all © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Aquaculture Research, 2015, 46, 346–357 treatments, net returns above both the TC and TVC were significantly better in fish fed diet D25. At increased inclusion levels of C. nilotica, net returns above TVC and TC reduced with increasing feeding duration because of lower economic return from the fish associated with reduced harvest weight of fish. Economic returns in diet D100 the lowest probably because of high feed costs, low survival and poor growth response. The enterprise budget analysis of diets in this study indicate that the best economic benefits when feeding fish with diets formulated using C. nilotica was realized at 25% inclusion level of C. nilotica. Nevertheless, it is economically feasible to culture O. niloticus based on diets formulated using C. nilotica inclusion levels of up to 75%. In conclusion, the results of this study suggest that it is possible to substitute up to 75% of the FM by C. nilotica in diets with low fish meal content for O. niloticus without affecting growth performance, feed and nutrient utilization as well as economic benefits. For test ingredients to be suitable as protein sources for feed formulation, they need to contain high protein content, excellent amino acid profiles, high nutrient digestibility and should have relative low price comparedwith other protein ingredients. These properties were perfectly satisfied by C. nilotica making it a suitable alternative protein source to fish meal. Our current finding lends credence to the continued research into areas of utilization of alternative locally available proteins sources in place of fishmeal as protein sources in improving aquaculture in areas with abundant C. nilotica. Furthermore, trials on culture of C. nilotica may suggest ways of commercial production of the ingredient for fish feed formulation. Acknowledgements We would like to acknowledge the financial support granted by Pond Dynamics/Aquaculture Collaborative Research Support Program (PD/A CRSP), partially funded by the United States Agency for International Development (USAID) under Grant No. LAG-G-00-96-90015-00. The New Partnership for Africa’s Development (NEPAD) initiative, which resolved that important aquaculture issues including research initiatives into fish nutrition require sound application in local contexts rather than high level innovations, motivated this research. © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Utilization of Caridina nilotica Mugo-Bundi et al. References Abdelghany A.E. (2000) Optimum dietary protein requirements for Oreochromis niloticus L. fry using formulated semi-purified diets. In: Tilapia Aquaculture in the 21st Century. Proceedings from the 5th International Symposium on Tilapia Aquaculture, Rio de Janeiro, Brazil, 3–7 September 2000, Vol. 1 (ed. by K. Fitzsimmons & J.C. Filho), pp. 101–108. Abdelghany A.E. (2003) Partial and complete replacement of fish meal with gambusia meal in diets for red tilapia (Oreochromis niloticus 9 O. mossambicus). Aquaculture Nutrition 9, 145–154. Abdel-Tawwab M., Khattab Y.A.E., Ahmad M.H. & Shalaby A.M.E. (2006) Compensatory growth, feed utilization, whole body composition and haematological changes in starved juvenile Nile tilapia, Oreochromis niloticus (L.). Journal of Applied Aquaculture 18, 17–36. Abdel-Tawwab M., Mohammad H., Ahmad M.H., Khattab Y.A.E. & Shalaby A.M.E. (2010) Effect of dietary protein level, initial body weight, and their interaction on the growth, feed utilization, and physiological alterations of Nile tilapia, Oreochromis niloticus (L.). Aquaculture 298, 267–274. Al-Hafedh Y.S. (1999) Effects of dietary protein on growth and body composition of Nile tilapia, Oreochromis niloticus L. Aquaculture Research 30, 385– 393. AOAC (1990) Official Methods of Analyses (15th edn) (ed. by K. Helrich). Association of Official Analytical Chemists Inc., Arlington, VA, USA, 1298pp. Balirwa J.S., Chapman C.A., Chapman L.J., Cowx I.G., Geheb K., Kaufman L., Lowe-McConnell R.H., Seehausen O., Wanink J.H., Welcomme R.L. & Witte F. (2003) Biodiversity and fisheries sustainability in the Lake Victoria basin: an unexpected marriage? BioScience 53, 703–716. Boyd C.E. (1984) Water Quality in Warm water Fishponds. Auburn University Agriculture Experimental Station, Auburn, AL, USA. Budeba Y.L. (2007) Contribution of Caridina nilotica (Roux) in the Dagaa Fishery in Lake Victoria, Tanzania. Aquatic Ecosystem Health and Management 10, 381–391. Budeba Y.L. & Cowx I.G. (2007) The role of the freshwater shrimp Caridina nilotica (Roux) in the diet of the major commercial fish species in Lake Victoria, Tanzania. Aquatic Ecosystem Health and Management 10, 368–380. Chepkirui-Boit V., Ngugi C.C., Bowman J., Oyoo-Okoth E., Rasowo J., Mugo-Bundi J. & Cherop L. (2011) Growth performance, survival, feed utilization and nutrient utilization of African catfish (Clarias gariepinus) larvae cofed Artemia and a micro-diet containing freshwater atyid shrimp (Caridina nilotica) during weaning. Aquaculture Nutrition 17, e82–e89. 355 Utilization of Caridina nilotica Mugo-Bundi et al. Cowx I.G., Van der Knaap M., Muhoozi L.I. & Othina A. (2003) Improving fishery catch statistics for Lake Victoria. Journal of Ecosystem Health and Management 6, 299–310. Degani G., Viola S. & Yehuda Y. (1997) Apparent digestibility of protein and carbohydrate in feed ingredients for adult tilapia (Oreochromis aureus 9 O. niloticus). Israeli Journal of Aquaculture 49, 115–123. Dias J., Alvarez M.J., Diez A., Arzel J., Corraze G., Bautista J.M. & Kaushik S.J. (2005) Dietary protein sources affects lipid metabolism in the European seabass (Dicentrarchus labrax). Comparative Biochemistry and Physiology 142A, 19–31. El-Saidy D.M.S.D. & Gaber M.M.A. (2003) Replacement of fish meal with a mixture of different plant protein sources in juvenile Nile tilapia Oreochromis niloticus (L.) diets. Aquaculture Research 34, 1119–1127. El-Sayed A.F.M. (1998) Total replacement of fish meal with animal protein sources in Nile tilapia Oreochromis niloticus (L) feeds. Aquaculture Research 29, 275–280. Fagbenro O.A. (1998) Apparent digestibility of various legume seed meals in Nile tilapia diets. Aquaculture International 6, 83–87. Falaye A.E. & Jauncey K. (1999) Acceptability and digestibility by tilapia Oreochromis niloticus of feeds containing cocoa husk. Aquaculture Nutrition 5, 157–161. FAO. (2006) State of the World Aquaculture 2006. FAO Fisheries Technical Paper, Vol. 500, FAO, Rome. 134pp. Fasakin E.A., Serwata R.D. & Davies S.J. (2005) Comparative utilization of rendered animal derived products with or without composite mixture of soybean meal in hybrid tilapia (Oreochromis niloticus 9 Oreochromis mossambicus) diets. Aquaculture 249, 329–338. Fontainhas-Fernandes A., Gomes E., Reis-Henriques M.A. & Coimbra J. (1999) Replacement of fish meal by plant proteins in the diet of Nile tilapia: digestibility and growth performance. Aquaculture International 7, 57–67. Gaber M.M.A. (1996) Partial and complete replacement of fish meal by poultry by-product and feather meal in diets of Nile tilapia, Oreochromis niloticus. Annals of Agricultural Science 32, 203–214. Gatlin D.M., Barrows F.T., Brown P., Dabrowski K., Gibson T., Gaylord K., Gaylord T.C., Hardy R.W., Herman E., Hu G., Krogdahl A., Nelson R., Overturf K., Rust M., Sealey W., Stoneberg D. & Souza E.J. (2007) Expanding the utilization of sustainable plant products in aquafeeds: a review. Aquaculture Research 38, 551–579. Getabu A., Tumwebaze R. & MacLennan D.N. (2003) Spatial distribution and temporal changes in the fish populations of Lake Victoria. Aquatic Living Resources 16, 159–165. Hanley F. (1987) The digestibility of foodstuff and the effects of feeding selectively on digestibility determination in tilapia, Oreochromis niloticus (L). Aquaculture 66, 163–179. 356 Aquaculture Research, 2015, 46, 346–357 Hern andez C., Olvera-Novoa M.A., Hardy R.W., Hermosillo A., Reyes C. & Gonz alez B. (2010) Complete replacement of fish meal by porcine and poultry by-product meals in practical diets for fingerling Nile tilapia Oreochromis niloticus: digestibility and growth performance. Aquaculture Nutrition 16, 44– 53. Khattab Y.A.E., Ahmad M.H., Shalaby A.M.E. & AbdelTawwab M. (2000) Response of Nile tilapia (Oreochromis niloticus L.) from different locations to different dietary protein levels. Egypt Journal of Aquatic Biology and Fisheries 4, 295–311. Liti D.M., Waidbacher H., Straif M., Mbaluka R.K., Munguti J.M. & Kyenze M.M. (2006) Effects of partial and complete replacement of fresh water shrimp meal (Caridina nilotica, Roux) with a mixture of plant protein sources on growth performance of Nile tilapia (Oreochromis niloticus L.) in fertilized ponds. Aquaculture Research 37, 477–483. Maina J.G., Beames R.M., Higgs D., Mbugua P.N., Iwama G. & Kisia S.M. (2002) Digestibility and feeding value of some feed ingredients fed to tilapia, Oreochromis niloticus (L.). Aquaculture Research 33, 853–862. Matsuishi T., Muhoozi L., Mkumbo O., Budeba Y., Njiru M., Asila A., Othina A. & Cowx I.G. (2006) Are the exploitation pressures on the Nile perch fisheries resources of Lake Victoria a cause for concern? Fisheries Management and Ecology 13, 53–71. Munguti J.M., Waidbacher H., Liti D.M., Straif M. & Zollitsch W. (2009) Effects of substitution of freshwater shrimpmeal (Caridina nilotica Roux) with hydrolyzed feathermeal on growth performance and apparent digestibility in Nile tilapia (Oreochromis niloticus L.) under different culture conditions. Livestock Research for Rural Development 21, 1–13. New M.B. & Wijkstro€ om U.N. (2002) Use of Fishmeal and Fishoil in Aquafeeds: Further Thoughts on the Fishmeal Trap. FAO Fish Circ. No. 975, FAO, Rome, Italy. NRC (National Research Council). (1993) Nutritional Requirements of Fish. National Academy Press, Washington, DC, USA. Olivera-Castillo L., Pino-Aguilr M., Lara-Flores M., Granados-Purto S., Montero-Munoz J., Olvera-Novoa M.A. & Grant G. (2011) Substitution of fish meal with raw or treated cowpea (Vigna unguilata L Walp, IT86-D719) meal in diet for Nile tilapia (Oreochromis niloticus L.) fry. Aquaculture Nutrition 17, e101–e111. Pouomogne V., Takam G. & Pouemegne J. (1997) A preliminary evaluation of cacao husks in practical diets for juvenile Nile tilapia (Oreochromis niloticus). Aquaculture 156, 211–219. Pullin R.S.V. (1996) Word tilapia culture and its future prospects. In: The Third International Symposium on Tilapia in Aquaculture (ed. by R.V.S. Pullin, T.J. Lazard, M. Lengendre, & K.J. Amon), pp. 1–16. ICLARM, Manila, Philippines. © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Aquaculture Research, 2015, 46, 346–357 Rasowo J., Ngugi C.C. & Macharia M. (2008) Comparative study of the growth performance of African catfish Clarias gariepinus (Burchell) fingerlings fed diets based on fresh water atyid shrimp meal, dagaa fish meal and soybean meal protein sources. East African Journal of Pure and Applied Science 1, 1–6. Riche M., Trottier N.L., Ku P.K. & Garling D.L. (2001) Apparent digestibility of crude protein and apparent availability of individual amino acids in tilapia (Oreochromis niloticus) fed phytase pretreated soybean meal diets. Fish Physiology and Biochemistry 25, 181–194. Richie M. & William T.N. (2011) Fish meal replacement with solvent extracted soybean meal or soy protein isolate in a practical diet formulation for Florida pompano (Trachinotus carolinus, L) reared in low salinity. Aquaculture Nutrition 17, 368–379. Richter N., Siddhuraju P. & Becker K. (2003) Evaluation of nutritional quality of moringa (Moringa oleifera Lam) leaves as an alternative protein source for Nile tilapia (Oreochromis niloticus L.). Aquaculture 217, 599–611. Soivio A., Niemisto M. & Backstrom M. (1989) Fatty acid composition of Coregonus muksun Pallas: changes during incubation, hatching, feeding and starvation. Aquaculture 79, 163–168. © 2013 John Wiley & Sons Ltd, Aquaculture Research, 46, 346–357 Utilization of Caridina nilotica Mugo-Bundi et al. Sugiura S.H., Dong F.M., Rathbone C.K. & Hardy R.W. (1998) Apparent protein digestibility and mineral availabilities in various feed ingredients for salmonid feeds. Aquaculture 159, 177–202. Tacon A.G.J. (1990) Standard Methods for the Nutrition and Feeding of Farmed Fish and Shrimp. Argent Laboratories Press, Washington DC, USA, 454pp. Tacon A.G.J. (1993) Feed Ingredients for Warmwater Fish: Fish Meal and Other Processed Feedstuffs. FAO Fish Circ. No. 856, FAO, Rome, Italy. Tacon A.G.J. & Jackson A.J. (1985) Utilization of conventional and non-conventional protein sources in practical fish feeds. In: Nutrition and Feeding of Fish (ed. by C.B. Cowey, A.M. Mackie, & J.G. Bell), pp. 119–145. Academic Press, London. Tacon A.G.J. & Metian M. (2008) Global overview on the use of fish meal and fish oil in industrially compounded aquafeeds: trends and future prospects. Aquaculture 285, 146–158. Viveen W.J.A.R., Richter C.J.J., van Oordt P.G.W.J., Janssen J.A.L. & Huisman E.A. (1985) Practical Manual for the Culture of African Catfish Clarias gariepinus. The Netherlands Ministry for Development Co-operation, Section for Research and Technology, The Hague, The Netherlands. 122pp. 357