The mechanism underlying the decrease in minute ventilation (VE) observed under halothane anesthesia was investigated in nine spontaneously breathing dogs. Anesthesia was induced with pentobarbital sodium and was maintained with halothane. Inspired fraction of halothane (F1hal) was increased every 30 min, from 0.005 to 0.02. VEdecreased from 8.1 ± 0.9 to 4.8 ± 0.41 min−1(P< 0.001), as F1hal increased from 0 to 0.02. This resulted from a decrease in both mean inspiratory flow (VT/T1) and the duty ratio (T1/TTOT). Transdiaphragmatic pressure (Pdi) and the integrated electrical activity of both hemidiaphragms (Edi) were measured during normal breathing, and during breathing against closed airways (P0di, E0di), in order to obtain an index of the inspiratory neuromuscular output of the diaphragm. With increasing F1hal, there was a significant decrease in Pdi, P0di, Edi, and E0di. The authors measured Pdi and Edi generated during supramaximal stimulation of the two phrenic nerves (P5di, E5di) at frequencies of 10,20,50, and 100 Hz, in order to eliminate in this decrease the role played by a decrease in the neural drive to breathing. P5di and E5di decreased significantly with increasing F1hal, and had not returned to the control values 30 min after discontinuation of halothane administration. The authors conclude that, in pentobarbital-anesthetized dogs, halothane is responsible for a diaphragmatic dysfunction, which may be located either at the neuromuscular junction, on the contractile processes of the muscle, or on both, and for a decrease in the activation time of the inspiratory muscles. Both of these effects contribute to the decrease in VEobserved under halothane anesthesia.