To measure the potency of pancuronium at the diaphragm and adductor pollicis in infants and children, train-of-four stimulation was applied to the ulnar and phrenic nerves under N2O-halothane anesthesia. The force of contraction of the adductor pollicis was measured and compared with the diaphragmatic electromyogram (EMG). Cumulative dose response curves were determined for pancuronium in 18 patients divided equally into three age groups: 0–1 yr, 1–3 yr, and 3–10 yr. The potency of pancuronium at both muscles decreased with increasing age (P < 0.05), while the adductor pollicis:diaphragm potency ratio remained constant. The mean doses (±SEM) required to depress adductor pollicis first twitch responses by 90% (ED90) were 42 ± 3.3 μg/kg in the 0–1-yr group, 47 ± 4.2 μg/kg in the 1–3-yr group, and 62 ± 4.1 μg/kg in the 3–10-yr group. Corresponding figures for the diaphragm were 70 ± 4.3 μg/kg, 81 ± 5.1 μg/kg, and 101 ± 4.4 μg/kg respectively. The ED90 ratios (diaphragm ED90/adductor pollicis ED90) in the three age groups were 1.69 ± .07, 1.75 ± .14, and 1.64 ± .09, respectively. These results are consistent with similar rates of maturation of the diaphragm and the adductor pollicis muscles in infancy and childhood. Thus, train-of-four monitoring of the adductor pollicis is likely to overestimate the degree of neuromuscular blockade of the diaphragm in pediatric patients.