We performed this study to assess the recovery period after neuromuscular blockade by electromyographic F-wave analysis, a method that supplies more information about more proximal parts of the motor system than conventionally used methods, e.g., mechanomyography (MMG). In 20 neurosurgical ASA physical status I or II patients anesthesia was induced and maintained with IV fentanyl and midazolam. Patients were randomly assigned to receive either 0.25 mg/kg mivacurium (MV group, n = 10) or 0.1 mg/kg pancuronium (PC group, n = 10) intraoperatively. MMG monitoring of the adductor pollicis muscle was performed continuously. F waves were recorded at the abductor pollicis muscle of the contralateral hand at train-of-four (TOF) ratios of 0.1, 0.25, 0.5, 0.7, 0.75, 0.8, 0.85, 0.9, and 0.95. Recovery of F-wave amplitudes after neuromuscular blockade with pancuronium was significantly slower compared with mivacurium (P = 0.004) during the clinically important recovery period defined by MMG TOF ratios from 0.7 to 0.95. This electrophysiologic finding suggests a differential recovery of the motor system after administration of pancuronium and mivacurium not detected by MMG.