Three anesthetics (pentobarbital, halothane, and isoflurane) were studied in six mongrel dogs to systematically compare their effects on chest wall function during spontaneous breathing. Each dog received each anesthetic on separate occasions. Electrical activities of several respiratory muscles were measured with chronically implanted electrodes, and chest wall motion was assessed by high-speed three-dimensional computed tomography scanning. Phasic expiratory muscle activity was markedly depressed by volatile anesthetics halothane and isoflurane compared with pentobarbital. In contrast, inspiratory activity in parasternal intercostal muscles was relatively well preserved during anesthesia with these volatile agents. The contribution of expiratory muscles to tidal volume was diminished during halothane and isoflurane compared with pentobarbital anesthesia. As anesthesia was deepened, expiratory muscle activity was unchanged during pentobarbital anesthesia, enhanced in some dogs during isoflurane anesthesia, and remained absent during halothane anesthesia. Activity in parasternal intercostal muscle was depressed as inspired concentration of halothane or isoflurane was increased, whereas diaphragmatic activity was unchanged. Depression of expiratory muscle activity by halothane persisted when breathing was stimulated by positive end-expiratory pressure, with significant mechanical consequences for chest wall configuration. Many of these findings are in contrast with previous observations in humans and suggest that the dog is not a suitable model for the study of the effects of anesthetic drugs on the pattern of human respiratory muscle activity.