In ten patients anesthetized with halothane and nitrous oxide,d-tubocurarine or pancuronium was infused continuously for 80 minutes to produce a constant 90 per cent depression of twitch tension. For the first 30-minute period, 8.2 ± 0.32 and 1.09 ± 0.06 (mean ± SE) mg/m2ofd-tubocurarine or pancuronium, respectively, were required, and for the last 30-minute period (50 to 80 minutes of infusion) 2.1 ± 10.10 and 0.41 ± 0.02 mg/m3were required, giving potency ratios of 8.2/1.1,i.e., 7.4, and 2.1/0.41,i.e., 5.1. The difference in potency ratios in the first and last 30-minute periods implies that potency values determined by single-injection techniques inadequately describe the relative requirements for sustained paralysis. The mean ratios between that amount of relaxant representing tissue uptake and that amount representing metabolism and renal excretion during the first 30 minutes were 3.06 ± 0.28 ford-tubocurarine and 1.63 ± 0.19 for pancuronium. The significantly higher ratio ford-tubocurarine implies that its tissue uptake relative to metabolism and renal excretion is greater than that for pancuronium. The higher ratio ford-tubocurarine may be partly explained by its greater binding to plasma and tissue proteins.