The interaction between the pulmonary afferents (PA) and the pneumotaxic center (PC) in control of respiratory pattern was studied in lightly anesthetized paralyzed cats before and after bivagotomy or lesions of the PC using inflations controlled by the onset or cessation of phrenic nerve discharge, i.e., cycle-triggered inflations. This interaction was also studied using electrical stimulation of the central stumps of cut vagi. Introduction of a delay between inspiratory onset and the commencement of an inflation at constant flow and duration resulted in increases of the durations of inspiration (T1) and expiration (TE) and amplitude of the integrated phrenic nerve discharge (A). The lung volume at inspiratory cutoff, i.e., the volume threshold, increased markedly as T1 increased. There were linear relationships between T1 and TE and between T1 and A. At constant alveolar CO2 and tidal volume, the quantitative effects of delay were dependent on the rate of inflation; i.e., when the flow increased, the volume threshold for a given T1 decreased. Bilateral vagotomy abolished the effects of delay and flow. PC lesions, which resulted in apneusis when the cycle-triggered inflations were stopped, produced the following changes compared to the delay effects seen in intact cats: a) the volume threshold for zero delay doubled and its rate of decrease with increased T1 was significantly smaller, and b) the change in TE for a given change in T1 was reduced markedly. Introduction of a delay between inspiratory onset and the start of electrical stimulation of the afferent vagi resulted in effects similar to those seen for delays in cycle-triggered inflations. The T1-TE relationship remained linear when the stimulus trains ended with inspiratory cessation. These results suggest that: a) the inspiratory cutoff mechanism is responsive to the rate, as well as the level, of lung inflation; b) all of the lung volume information affecting inspiratory cutoff in paralyzed cats is carried via the vagi; c) an intact PC is necessary for the generation of a normal time dependence of the volume threshold for inspiratory cutoff; d) the PC plays an important role in matching TE to T1 when the latter changes. For inflations and vagal stimulations applied during expiration, with introduction of a delay between inspiratory cessation and the start of cycle-triggered inflation or vagal stimulation, the results indicated that the expiratory cutoff mechanism has an irrevocable phase of 300-450 ms.