A modification of the single-breath nitrogen closing volume (CV) test allows measurement of closing capacity (CC) during general anesthesia. In tlie modification, inspiration and expiration are mechanically produced by a hydraulically powered cylinder. For 14 awake, normal subjects, results of the CV test performed using this mechanical method differed than those obtained following spontaneous inspiration and expiration. Mean (±SE) CC's were 2.25 (±0.15) and 2.421 (±0.14) (P< 0.01) using spontaneous and mechanical methods, respectively. The slopes of Phase III of the CV traces were 2.24 (±0.27) and 2.66 per cent N2/I (±0.32) (P< 0.01), respectively. To eliminate differences due to measurement technique, the modified CV test was used both before and during anesthesia with halothane in 70 per cent N2in 11 normal, supine, spontaneously breathing subjects. CC's were 1.891 (±0.16) before and 1.841 (±0.15) during anesthesia (P> .5). Mean functional residual capacities (FRC) by the closed-circuit helium method were 1.771 (±0.15) before and 1.451 (±0.17) during anesthesia (P< .001). With CC unchanged and FRC decreased following induction. CC/FRC increased from 1.07 (±0.08) to 1.37 (±0.11) (P< .005), suggesting increased small-airway closure during anesthesia.