We explored the possibility that glucocorticoid-induced muscle weakness and atrophy resulted from impaired muscle membrane excitability. Male Sprague-Dawley rats received intramuscular injections of dexamethasone, cortisone acetate (equivalent anti-inflammatory doses), or saline for up to 28 days. Temporal patterns of change in muscle mass, twitch and tetanic tension, and membrane potential, cable parameters, and excitability were studied in vitro in the extensor digitorum longus (EDL), soleus (SOL), omohyoid (OMO), caudofemoralis (CF), and the sternomastoid muscles (membrane potential only). the membrane properties of EDL fibers were also studied in vivo (pentobarbital anesthesia). The relative severity of atrophy was OMO greater than CF greater than EDL greater than SOL. Reduction in twitch or tetanic tension never preceded atrophy. The twitch and tetanic tension (per g muscle) increased with glucocorticoid treatment. There were no significant changes in the time course of the twitch or tetanus. Dexamethasone produced more severe atrophy and force reduction than did cortisone acetate. Glucocorticoid treatment produced a depolarization of EDL muscle fibers measured in vitro at 23 degrees C, but this did not appear to be physiologically significant because EDL fibers studied in vivo were not depolarized and had normal action potential amplitudes and thresholds. Glucocorticoid treatment did not change the membrane resistance or capacitance. We conclude that glucocorticoid treatment did not produce muscle weakness by impairing sarcolemmal excitability or excitation-contraction coupling, but that the weakness resulted from muscle atrophy.