1. Three isoenergetic test diets containing 320, 430 and 530 g protein/kg, with (experimental) and without (control) inclusion of an anabolic steroid, ethylestrenol, were given to rainbow trout (Salmo gairdneri) of mean initial weight 27 g, for 60 d.
2. After 60 d, all trout groups were given an identical, steroid-free, commercial diet (410 g protein/kg) for a further 30 d, to observe withdrawal effects on growth.
3. The weight and length of trout given the 430 and 530 g protein/kg control diets were significantly greater after 60 d than those given the 320 g protein/kg control diet.
4. Inclusion of steroid enhanced the weight and length of trout given the 320 and 430 g protein/kg experimental diets, exerting a preferential effect on weight as opposed to length.
5. After steroid withdrawal, a significant difference between the weight and length of the 320 g protein/kg control and experimental groups was still apparent.
6. An increase in dietary protein increased the food conversion efficiency, and decreased the protein efficiency ratio. Inclusion of steroid increased both these factors over the respective controls.
7. Protein assimilation decreased, and faecal nitrogen content increased with increasing dietary protein. In trout given steroid, protein assimilation and apparent digestibility was higher, and facecal N content, lower than the controls.
8. Muscle protein increased with increasing dietary protein, and was higher in trout given steroid.
9. Relative liver weight increased with increasing dietary protein. Inclusion of steroid resulted in a reduction in relative liver and gut weight, and an increase in kidney weight.
10. It is concluded that ethylestrenol promotes both the growth and efficiency of nutrient utilization of trout. The magnitude and duration of these effects are a function of the dietary protein level.