The measurement of force of contraction of the adductor pollicis muscle following supramaximal stimulation of the ulnar nerve has become a standard method to assess the effect of neuromuscular blocking drugs. However, the diaphragm is regarded as resistant to these drugs, and considerable residual respiratory power might still be present after total block of adductor pollicis function. To quantify this differential effect, train-of-four stimulation was applied to the ulnar and the phrenic nerves in patients under N2O-halothane anesthesia. The force of contraction of the adductor pollicis muscle was measured with a force-displacement transducer and compared with the diaphragmatic electromyogram (EMG). Pancuronium cumulative dose-response curves for both muscles were determined in 10 ASA Class I adults. The mean dose (±SEM) required to depress adductor pollicis and diaphragm responses to first twitch stimulation (ED50) was 29.5 ± 3.5 μg/kg and 59.5 ± 7.0 μg/kg, respectively. Corresponding values for ED90were 45 ± 5 μg/kg and 95 ± 11 μg/kg, respectively, indicating that the diaphragm required approximately twice as much pancuronium as the adductor pollicis for an identical block. At 90% adductor pollicis block, the diaphragm was only 24 ± 4% blocked. It is concluded that the adductor pollicis response might underestimate the degree of diaphragmatic relaxation. On the other hand, the administration of pancuronium in a dose sufficient to produce total paralysis might result in the inability to antagonize neuromuscular block in all muscles.