An Evaluation of Three Fish Diets Formulated from Locally Generated Agricultural By-products for Catfish ( Heterobranchus bidorsalis : Geoffroy St . Hilaire , 1809 ) culture

quaculture accounts for less than 10% of total fish production in Nigeria (Egwui, 1986; Aguigwo, 1993; Absalon and Omenaihe, 2000; Adekoya, 2001; Nwosu et al., 2002; Nwosu et al., 2009). Various workers (Huisman, 1986, Middendorp, 1993; Chidobem and Anibeze, 1994; Eze et al., 2000; Nwosu et al., 2002 and Nwosu et al., 2009) have advocated that aquaculture could bridge the gap and demand supply of fish in Nigeria. One major constraint to the development of aquaculture, however including other constraints (Okeke et al., 2016) is the prohibitive high cost of extruded fish feeds in the country. Some resource poor farmers often resort to the use of fish feeds of doubtful quality which often results in low yield. Recent studies (Nwosu et al., 2009; Okeke et al., 2016) showed that Clarias garipinus (Burchell: 1822) and Heterobranchusspp. are preferred in Nigeria because they are hardy, grow fast and resistant to diseases. Adequate knowledge of these species and their nutritional requirements etc. is therefore, necessary in order to sustain, their continuous production.


INTRODUCTION
quaculture accounts for less than 10% of total fish production in Nigeria (Egwui, 1986;Aguigwo, 1993;Absalon and Omenaihe, 2000;Adekoya, 2001;Nwosu et al., 2002;Nwosu et al., 2009).Various workers (Huisman, 1986, Middendorp, 1993;Chidobem and Anibeze, 1994;Eze et al., 2000;Nwosu et al., 2002 andNwosu et al., 2009) have advocated that aquaculture could bridge the gap and demand supply of fish in Nigeria.One major constraint to the development of aquaculture, however including other constraints (Okeke et al., 2016) is the prohibitive high cost of extruded fish feeds in the country.Some resource poor farmers often resort to the use of fish feeds of doubtful quality which often results in low yield.Recent studies (Nwosu et al., 2009;Okeke et al., 2016) showed that Clarias garipinus (Burchell: 1822) and Heterobranchusspp.are preferred in Nigeria because they are hardy, grow fast and resistant to diseases.Adequate knowledge of these species and their nutritional requirements etc. is therefore, necessary in order to sustain, their continuous production.
Various research work (Alfred-Ockiya et al., 1995;Eze et al., 2000;Nwosu et al., 2002;Jamabo and Alfred-Ockiya, 2008;Nwosu et al., 2009 andOkeke et al., 2016) have aptly demonstrated the growth response of cat fishes to agricultural by-products such as proteins sources.This, therefore, formed the pivot of the present research work viz: to evaluate three fish diets formulated from locally generated agricultural byproducts for catfish culture.

MATERIALS AND METHODS
The research work was carried out for a period of 56 weeks.There were three treatments namely: T1 (fish meal, soybean meal and yellow maize), T2 (cow blood meal, soybean meal and yellow maize), T3 (cassava peels, soybean meal and yellow maize) and the experimental fish.

ABSTRACT
The growth rate, economic benefits and other yield characteristics of Heterobranchus bidorsalis (Geoffroy St. Hilaire, 1809) fed with three diets formulated from locally generated low-cost agricultural by-products were evaluated in an experiment that was carried out for a period of 56 weeks.The best growth and profitability were obtained in T1 (Fish meal soybean meal and yellow maize followed by T2 (cow blood, soybeans meal and yellow maize) and T3 (cassava peels, soybean meal and yellow maize).Evaluation of growth parameter and food utilization of the fish fed on the three experimental diets was based on Mean Weight Gain (M.W.G.), Food Conversion Ratio (F.C.R.), and Survival Rate.There was significant difference (p<0.05) in M.W.G., M.F.W., F.C.R and Net Revenue realized from the sale of fish produced from the three treatments (T1, T2 and T3).Fish fed T1 had the highest M.W.G. (2,353.30g) and conversely the least F.C.R. (1.80) while fish fed T2 had a M.W.G. of 2,030.40gand F.C.R. of 2.80 while fish fed T3 had a M.W.G. of 1,261.15gand F.C.R. of 3.85.However, there was no significant difference (p>0.05) in survival rate.The economy of food conversion was best in T1.Net Revenue realized from the sale of fish was highest in T1 (N105, 764.50) followed by T2 (N74,440.50)and T3 (N39,438.50).
Each treatment was replicated 4 times, collection and stocking of fish fingerlings.A total of 1440 fingerlings of Heterobranchus bidorsalis with an initial mean weight of 4.50g0.02g and mean standard length of 7.30 cm 0.04 cm were stocked at the rate 120 fingerlings in 12 homestead concrete fish ponds 60m 3 owned by Anambra State Agricultural Development (ADP) contact farmers in the four Agricultural Zones viz: Aguata, Onitsha, Anambra and Awka (Fig. 1).The fingerlings were prior to stocking conditioned and acclimatized by introducing small quantities of the pond water into the plastic container holding the fish fingerlings for a period of 30 minutes.The fish ponds were prepared for stocking with fish fingerlings using methods adopted by Nwosu et al. (2009).The ponds were each fertilized with a 1:1 mixture of poultry dropping and cow dung at the rate of 9kg per pond to generate phytoplankton and zooplankton.

Pond water parameters
Pond water parameters (Table 1) were taken fortnightly by measuring pH, temperature and turbidity.Surface water temperature was measured with the aid of mercury in glass bulb thermometer at a depth of 25mm.pH was measured with the aid of a battery operated field pH meter while turbidity was measured with the aid of a 25cm Secchi Disc as stipulated by AOAC (1990) and Nwosu et al. (2009), while nitrate-nitrogen concentration was determined spectrophotometrically.Dissolved oxygen was determined using the Winkler method

Preparation of fish feed
The three fish feeds were compounded to contain 35% crude protein and the fish were fed 5% of their body weight.The level of hydrocyanic acid (HCN) in the cassava peels was reduced following methods adopted by Tewe (1982) and Nwosu et al. (2009): by sun-drying before being incorporated into the fish feed.The compounded fish feeds were supplemented with palm oil to ensure that they were 150caloric (Tewe, 1982;Nwosu et al., 2009).The experimental fish were fed twice daily (7am and 6pm) and sampling was done fortnightly to monitor fish growth.
The growth performance and food utilization parameters of H. bidorsales (Table 2) were calculated using methods presented in Nwosu et al. (2009) and Okeke et al. (2016).
1. Food conversion ratio -This was determined as the ratio of food consumed by fish to the weight gain by fish and expressed as

Economic Analysis
Cost-benefit Analysis (Table 3) was based on methods adopted by Nwosu et al. (2002) and Okeke et al. (2016).The following parameters were analysed.
1. Cost/kg feed (N) = properties of each ingredient in the diet  cost/kg of the increment.2. Cost of total feed consumed per fish (N) = total feed consumed  cost/kg feed.3. Cost/kg weight gain = cost/kg  feed conversion 4. Revenue (total cost or net cost) cost/kg fish  mean weight gain 5. Gross margin = revenuescost of total feed consumed.

Statistical Analysis
The growth performance of H. bidorsalis as affected by the three feeds were computed at the end of the research work and subjected to one-way Analysis of Variance (ANOVA) and least significant difference (LSD) test at 5% level of significance.

RESULTS AND DISCUSSION
The growth performance and food utilization parameters of H. bidorsalis fed three diets viz, T1, T2 and T3 are presented in Table 1.The highest growth response (mean weight gain) of 2,352.30gwas attained in T1 followed by T2 (2,025.90g) and T3 (1,261.15g) which varied significantly (p<0.05).
The economy of feed conversion consistently was most efficient (1.80) in T1 (Fish meal and yellow maize followed by T2 (cow blood, soybean meal and yellow maize with a food conversion ratio (F.C.R.) of 2.80 and 3.85 in T3.The difference in growth performance and food utilization of H. bidorsalis fed with three experimental diets (T1, T2 and T3) could be attributed to the different sources of protein in the diets.Work done elsewhere (Eze et al., 2000;Nwosu et al., 2002;Nwosu et al., 2009) demonstrated that fish growth is proportional to the amount of protein in the diet.It is noteworthy also that although all three fish diets used contained protein only T1 contained fish protein which is a high grade protein (Egwui, 1986;Ockiya et al., 1995;Eze et al., 2000;Nwosu et al., 2002 andOkeke et al., 2016).
The lowest growth response was obtained in T3 which included cassava peels, soybean meal and yellow maize.Work done elsewhere (Nwosu et al., 2009) has shown that as in the case of poultry and other monogastric animals certain precautionary measures need to be taken to guarantee satisfactory performance of the fish stock fed on cassava meal.It has been demonstrated for instance that cassava can affect mineral imbalance in monogastric animals such as pigs (Tewe, 1982).This condition however, can be prevented by including zinc salts in the diets (Maust et al., 1972 andNwosu et al., 2009).

Percentage Survival
Percentage survival of the experimental fish was generally high although not significantly different (p>0.05) for all the treatments (95.83%, 94.60% and 9.25%) in T1, T2 and T3 respectively.The high survival rate of H. bidorsalis recorded in the present work may be attributed to the high ecological tolerance of the clariid and proper pond management.

Economic Analysis
A simple cost-benefit analysis (Table 2) obtained from total weight of fish produced, gross revenue, total cost of product and net revenue showed that net profit from T1 was N105,764.50;N74,440.50 from T2 and N39,438.50 from T3.

Pond Water Parameters
Pond water quality (Table 3) was not adversely affected in the course of the research work.The mean values of water quality parameters did not vary significantly (p>0.05) and fell within the acceptable range for fish culture in the tropics (AOAC, 1990).

CONCLUSION
The results of this research work showed that both fish yield and net profit were higher in T1 followed by T2 and T3.The high level of fish production in T1 tends to suggest that the inclusion of fish meal and soybean meal in fish diets enhances food utilization, growth and weight gain in culture fish.
The lowest value of fish production obtained in T3 may be attributed to the incorporation of cassava peels which may alter nutrient balance in fish feeds.This result tends to suggest an optimum, minimum and maximum inclusion/substitution levels for cassava in fish feed formulation.These observations notwithstanding the beneficial effects of cassava peels and the other two agricultural by-products used in the present research work cannot be underestimated in the quest to reduce cost of cat fish production in the country.