Thesis(M.S.)--Auburn University, 2005. / Abstract. Vita. Includes bibliographic references.

Thesis (M.S.)--University of North Carolina at Wilmington, 2004. / Includes bibliographical references (leaves : 50-54).

Lack of affordable feeds is one of the major constraints facing small-scale fish farmers in Tanzania. This study evaluated the suitability of moringa leaf meal (MLM), cassava leaf meal (CLM) and cassava root meal (CRM) as novel ingredients in Nile tilapia, Oreochromis niloticus diets. Each of the ingredients was processed in an attempt to remove the most significant antinutritional factor. A series of five experiments was conducted in a recirculation system using juvenile O. niloticus. The fish were fed isonitrogenous (30g 100g-1), isolipidic (10g 100g-1) and isoenergetic (18 kJ g-1) diets containing graded levels of the processed ingredients to their apparent appetite but not exceeding 10% of their body weight for a period of 8 weeks. Processing led to the removal of 0.3% of saponin from MLM and 60% and 90% of hydrogen cyanide from CLM and CRM respectively. The contents of other inherent antinutritional factors such as phenols, tannins and phytic acid were little affected. Processed MLM, CLM and CRM had 31.1/29.0/1.5g 100g-1 of crude protein, 5.9/10.2/2.4g 100g-1 of crude fibre and 20.1/19.7/15.8k Jg-1 of gross energy. The content of sulphur amino acids was higher in CLM (0.47%) than in MLM (0.23%). Digestibe protein and digestible energy was higher in MLM (25.71g 100g-1/15.44kJ g-1) than in CLM (12.71g 100-1/9.16kJ g-1). CRM had a digestible energy content of 13.5kJ g-1. Inclusion of either of the leaf meals, even at the lowest level of 15g 100g-1 of total dietary protein, led to a significant reduction in feed intake, growth and feed utilisation. Liver and small intestine did not show any histopathological changes which could be linked to dietary treatment. Conversely, cassava root meal could replace up to 75% of wheat meal in the diet without significantly affecting performance. The performance of leaf meals was marginally improved by a combination of blending and feeding stimulants, whereby a blend containing 1 part MLM and 2 parts CLM could provide up to 20g 100g-1 of dietary protein without significantly reducing performance. Biological and economic performance of practical diets containing 30-50g 100g-1 of dietary protein from moringa and cassava blends (LMB) with feeding stimulants was significantly lower than a fishmeal-meal based diet (FM) but comparable to a soybean meal-based diet (SBM). The suitability of MLM and CLM as novel protein sources in O. niloticus diets will depend on 1) improving reduction/removal of inherent antinutritional factors in MLM and CLM as well as improving digestibility of CLM. On the other hand, the suitability of CRM as a carbohydrate energy source will depend on the availability of cost effective protein sources due to its low protein content.