To test the hypothesis that cortisol has a regulatory role in fish muscle glycogenesis post-exercise, rainbow trout were treated 1 h prior to exercise with either saline (control) or metyrapone (2-methyl-1, 2-di-3-pyridyl-1-propanone) to block cortisol synthesis. Following exercise (time 0), half of the metyrapone-treated fish received a single injection of cortisol, to mimic the post-exercise rise usually observed. Muscle glycogen and the relative activities of glycogen phosphorylase a (Phos a) and glycogen synthase I (GSase I), regulatory enzymes for glycogen resynthesis, were monitored 4 h post-exercise. Metyrapone treatment succeeded in blocking the post-exercise rise in plasma cortisol (17+/-2 vs 118+/-13 ng ml(-1) in controls at time 0), and cortisol injection resulted in a larger and more prolonged cortisol increase than in controls (159+/-22 vs 121+/-14 ng ml(-1) in controls at 1 h). Muscle glycogen was completely restored in the metyrapone-treated fish within 2 h after exercise (8.3+/-0.6 vs 8+/-0.7 micromol g(-1) pre-exercise), only partially restored in control fish at 4 h (5.4+/-01.4 vs 8.8+/-1.3 micromol g(-1) pre-exercise), and not at all in cortisol-treated fish (1.0+/-0.5 micromol g(-1) at 4 h). The rapid glycogen resynthesis in the metyrapone-treated fish was associated with a more rapid inactivation of Phos a and stimulation of GSase I compared to controls. In cortisol-treated fish, Phos a activity persisted throughout 4 h post-exercise; there was also a significant stimulation of GSase I activity. As a consequence of dual activation of Phos a and GSase I, glycogen cycling probably occurred, thus preventing net synthesis. This explains why the post-exercise elevation of cortisol inhibits net glycogen synthesis in trout muscle.