TIFICIAL respiration has become synonymous with the positive pressure respirator. These mechanical devices successfully ventilate patients with acute respiratory failure. However, when artificial ventilation is necessary for an extended time, alterations occur in the tracheobronchial tree and lung parenchyma; these include retained secretions, atelectasis, pneumonitis, and interstitial pulmonary edema. As obstructive or inflammatory lesions develop in the lungs there is an increase in the physiological dead space and the related ventilation requirements. Pontoppidan and co-workers report the ratio of dead space to tidal volume may become so large that no improvement in alveolar ventilation occurs with increased tidal volumes? They also found impaired oxygenation in all their patients. Experience indicates that pulmonary complications contribute significantly to the mortality and morbidity of neurosurgical patients, and that these complications are neither minimized nor avoided by our present ventilation techniques. In our search for an improved ventilation technique we have assumed that, if the normal inspiratory muscles could be electrically driven, the chest cage elevated, and the lung uniformly expanded by negative pressure, some reduction in the high pulmonary complication rate might occur. Little attention was given to electronic driving of the normal neuromuscular apparatus in the work of respiration until the reports by Sarnoff and co-workers in 1948. They demonstrated the feasibility of artificial diaphragmatic respiration by electrical phrenic