Mixed sex eyed rainbow trout Oncorhynchus mykiss embryos were obtained (Spring Lake Trout Farm, UT, USA) and incubated in California trays at the Ohio State University Aquaculture Laboratory (OSUAL). When yolk sac was determined to be almost completely absorbed, larvae (n=62) were randomly stocked to eighteen 15-liter tanks in a recirculating culture system. The experiment was set up in a randomized complete block design with each block being a shelf with six tanks. Six treatment groups were applied in triplicate with five experimental diets which were formulated and produced at OSUAL and a commercial diet. The protein sources of the diets varied and included 50% and 75% raw fish waste, 50% fish waste fermented with a consortium of bacteria, 50% fish waste fermented with a single bacteria strain, and a fishmeal control diet. Fish in each tank were fed three times per day at a restrictive feeding rate. Samples of whole tank biomass were taken monthly to readjust feeding rates and to calculate survival, individual fish mass, feeding coefficient ratio, and specific growth rate. The density of fish was cut to 40 fish per tank at 30 days of feeding (dof) and further to 20 fish per tank at 60 dof until an estimated mean individual fish mass of 10 g was achieved in each diet treatment. When final samples for each diet treatment were taken five fish were flash frozen whole in liquid nitrogen and stored at - 80 °C for proximate and lipid analysis.
The different diet treatments achieved a mean individual mass of 10 g between 92 and 111 dof for all treatment groups, except for the diet with 75% fish waste as the protein source which still had not achieved this endpoint by 122 dof (Fig. 1). Growth was highest in the commercial control, followed by the fishmeal control and then the fish waste diets. The individual mean mass of fish in the 75 % Raw fish waste diet was similar to the other fish waste diets after 30 dof but was lower than these diets by 61 dof. This trend continued to increase in magnitude as the experiment progressed. This appears to be a function of significant decreases observed feed conversion ratio recorded at 92 (1.51 ± 0.05) and 122 (3.87 ± 0.72) dof for this diet which were 1.4-3 folds higher than the other fish waste diets. Results obtained for survival, proximate analysis, and fatty acid analysis of total lipids will also be presented. Differences in survival, growth, and feed conversion ratio of 50% fermented fish waste diets compared to 50% non-fermented were not readily apparent. These results indicate that incorporation of 50% fish waste product, fermented or raw, as the protein source may be acceptable for Rainbow trout growth and survival. Project was funded by United States Department of Agriculture U.S.-Israel Binational Agricultural Research and Development Fund.