A DISSERTATION SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN FOOD SCIENCE OF THE SOKOINE UNIVERSITY OF AGRICULTURE. MOROGORO, TANZANIA. 2017A survey was conducted at Mkuranga and Rufiji districts in the Coastal Region of Tanzania. A total of 10 samples comprised of two from each of 5 varieties were subjected to three preparation methods (pounded cassava leaves as preparation method 1, boiled pounded leaves for 30 minutes as preparation method 2, and pounded leaves in hot water then boiled for 30 minutes as preparation method 3). These samples were analyzed for proximate components, β-carotene, vitamin C and cyanide content. The data was subjected to analysis of variance (ANOVA) to compare the effect of preparation methods, location and variety on the above mentioned components. Means were separated by DMRT and tested for significant by differences attributed to preparation methods and variety. The findings indicate that cassava leaves from Kiroba variety is mostly cultivated by farmers in the study area which was significantly attributed to its being tasty and high yield, following significance test by DMRT at p<0.05. Locational mean difference were subjected to t test (p< 0.05). Mpira variety was found to contain higher moisture content (82.40%) than other varieties (p<0.05). Kizimbani variety was found to contain higher protein (31.78% DM) than other varieties. Nyakamgile contained significantly higher ash content (10.48% DM) compared to other varieties. Crude fiber was significantly higher in Cheupe (25.13% DM) than other varieties all mean differences were significant at p<0.05 level. Crude fat was higher in Kizimbani variety (3.91% DM) while Mpira variety contained significantly higher carbohydrate (36.10% DM) content (p<0.05). The cyanide content on cassava leaves was found to be higher in Kiroba (185.51mgHCN/kg) DM and lower in Nyamkagile variety (54.17mgHCN/kg) DM, cyanide content for the preparation method 1 was 376.31mgHCN/kg DM while that of preparation method 2 was 12.37mgHCN/kg DM and preparation method 3 was 9.25mgHCN/kg DM and for which the difference was significant at p<0.05.Cheupe variety had significantly higher β carotene content while vitamin C content was significantly higher in the Kizimbani variety. The preparation methods had no effect on fat and protein content, but changes were observed by increased in moisture and ash content of cassava leaves prepared with method 2 and 3, while there was reduction in crude fiber, carbohydrate, β- carotene, vitamin C and cyanogenic potential levels during preparation method 2 and 3. The observed variability in cassava leaves composition was significantly attributed to preparation methods, varieties and location (p<0.05). In this study preparation methods and varieties had no effect on colour, taste, aroma and general acceptability. Preparation method 2 had more effect on aroma compared to cassava leaves prepared with method 3.The texture of cassava leaves was affected by the variety used while it will remain the same irrespectively of any preparation method used. Based on these finding, African preparation methods 2 and 3 reduce the cyanide content with considerable nutrient retention. There should therefore be no fear in utilization of cassava leaves for direct consumption.