A thesis submitted in partial fulfillment of the requirements for the degree of the Master o f Science, School of Environment, Resources and Development

Thesis (M.Sc.) - Asian Institute of Technology, 1989

Different nutritional inputs were used to culture monosex Nile tilapia ( Oreoch romisniloticus ) with the same initial. individual mean weight of 5. 1 g in 200 m2 earth ponds for eighteen weeks. Four treatments with three replications each were used with the same fish stocking rate of 600 fish/pond <1:> follows: Treatment 1 , duck/fish integration; Treatment 2, buffalo manure and inorganic fertilizers (urea and triple superphosphate) with rice bran a supplementary feed given daily a t 2% of fish bod y weight; Treatment 3 , pelleted feed given initially <1t 2%/body weight /day and later at 3% of fish body weight; Treatment 4, no nutritional inputs. Mean dissolved oxygen concentrations (DO ) fell into two groups, manure d ponds (Treatments 1 and 2 ) and unmanaged panels (Treatment 3 a nd 4 ) . DO in manured ponds was lower at dawn and higher in the after noon than in unmanaged ponds , and these trends were nearly the same level through out the experimental period . Phytoplankton and zooplankton biomass were low in unmannered ponds but high in manured ponds ; l1 owever , they were higher in ponds loaded with duck manure than buffalo manure. Treatment 1gave the best result in terms of mean growth rate, mean specific growth rate , mean individual fish weight and net fish yield followed in decreasing order by Treatment 2,3 and 4 . There was no significant differences in survival rate be t ween the treatments , except in the control treatment with only 49. 8% survival rate . Using gross margin analysis , the economic results showed that only Treatment 1 gave a positive value with the others negative . Integration of duck/fish was an not an economical method eventhough it gave the best fish yield because of the financial loss of the duck enterprise. Using pelleted feed was also an uneconomical method but it could be economic if the feed price were reduced. No pond inputs was not economic due to high mortality and poor fish growth . Only the treatment using buffalo manure , inorganic fertilizers and rice bran as pond inputs may be economic and this strategy show potential for small-sca1c farmers.