Airway occlusion pressure has been used in the past two decades for assessing output of the respiratory controller. It gives a measurement of a weighted sum of the effect of all respiratory muscles active at a given time and, unlike ventilation or tidal volume, does not depend on the resistance or compliance of the respiratory system. In anesthetized subjects or animals, it gives a tracing of the time course of respiratory neuromuscular output through the respiratory cycle, modified by elimination of most phasic vagal stretch receptor feedback and perhaps slightly by activation of some chest wall reflexes. The original postulate that an occluded inspiration would be isometric and the measured pressure free from losses due to force-length and force-velocity has been shown to be incorrect. The volume at which occlusion takes effect, distortions of the chest wall during the maneuver, tonic vagal input, and strength of the muscles must be taken into account when the data are interpreted. Brief occlusions [pressure at 0.1 s (P0.1)] are useful in measuring output in the very first part of inspiration in conscious subjects but must be treated with a great deal of caution. They are most reliable when end-expiratory volume remains constant and there are no important phase lags between flow and pressure. Allowance may be necessary for damping of the pressure signal on its passing through the compliant upper airway. Changes in P0.1 may often be due to changes in the shape of the driving pressure wave without a proportionate change in overall output. The technique remains useful when its limitations are recognized. Because of its simplicity, it can be easily and usefully applied to a range of clinical investigations.