In this study, the effects of temperature and different levels of recombinant human somatotropin, one of the growth agents, on growth performances and body composition of the rainbow trout (Oncorhynchus mykiss) were investigated. Two different temperatures (10.1 +/- 0.10 degrees C and 14.8 +/- 0.50 degrees C) and four different doses of the growth agent were added to the feed (0.0, 27.4, 54.8 and 82.2 mu g/g) in this experiment. The following results were obtained from the research conducted according to 2 x 4 x 2 randomized factorial experimental design which lasted for a total of 230 days (114 feeding days with growth agent). At the end of the experiment, weights of the fingerlings whose initial weights were 0.23 +/- 0.04 g, differed between applications. While the highest value was obtained as 17.51 +/- 0.56 g from the 27.4 mu g/g growth agent at higher temperature group, the lowest values were obtained as 7.05 +/- 0.06 g from the control group at normal temperature and 8.35 +/- 0.19 g from the 82.2 mu g/g growth agent normal water temperature group. The differences between water temperatures and levels of growth agents were found to be statistically significant (p < 0.01). In terms of feed conversion ratio (FCR), different values were obtained from the groups and the differences between them were also found to be statistically significant (p < 0.01). While the best feed conversion ratio was obtained as 1.43 +/- 0.24 and 1.43 +/- 0.23 g from the 27.4 and 54.8 mu g/g growth agents at normal temperature groups, the lowest value was obtained as 1.85 +/- 0.11 in the control group at higher temperature. From the survival rate (SR) point of view, the difference between the results obtained from the groups was found to be statistically significant in terms of water temperatures, but not significant in terms of the levels of growth agent (p < 0.05). As a result of proximate analyses of fish meat, the differences between water temperatures and levels of growth agents were found to be statistically significant in terms of crude protein and crude lipid values (p < 0.01). The crude protein ratio was found to be higher and the crude lipid ratio was found to be lower in the groups given the growth agent. After investigating the results obtained from other analyses at the end of the study, the carcass weight was found to be statistically significant (p < 0.01), the condition factor (CF) was not found to be significant, and values of hepatosomatic index (HSI) and viscerosomatic index (VSI) were found to be statistically significant (p < 0.01) in terms of water temperatures and levels of the growth agent. According to GC-MS (gas chromatography-mass spectrophotometer) tests, residue of the growth agent was found to be negative in the body of fries 15 days after addition to the feed was ceased.