In early phases of respiratory disease, patients are more likely to experience intermittent hypercapnia than a continuous increase in PCO2. The effect of intermittent arterial PCO2 elevation on subsequent breathing patterns is unclear. To examine this issue, a series of six ventilatory challenges (CH1-CH6), consisting of 2 min of breathing 5% CO2 in O2, followed by 5 min in room air (RA) were performed in 10 naive healthy subjects (age 12-39 yr). Minute ventilation (VE) increased from 11.9 +/- 1.0 (SE) l/min in RA to 27.6 +/- 3.0 l/min in 5% CO2 (P < 0.0005) in each of the six hypercapnic challenges. Respiratory rate increased from 21.3 +/- 2.6 breaths/min on RA to 29.6 +/- 3.9 breaths/min during CH1 (P < 0.05). However, respiratory rate consistently decreased with successive CO2 challenges (CH6: 21.5 +/- 2.6 breaths/min; P < 0.02). Thus, maintenance of VE was achieved by gradual increases in tidal volume with each of the first four consecutive CO2 challenges (CH1: 1.05 +/- 0.09 liters; CH4: 1.44 +/- 0.13 liters; P < 0.002). Similarly, the ratio of tidal volume to inspiratory time increased from CH1 (1.16 +/- 0.16 l/s) to CH6 (1.57 +/- 0.21 l/s; P < 0.001). These changes in ventilatory strategy were not observed when RA recovery periods were extended to 15 min in five subjects. We conclude that during repeated short hypercapnic challenges similar levels of VE are achieved. However, increased mean inspiratory flows are generated to maintain VE. We speculate that intermittent hypercapnia either modifies central controller gain or induces a long-term modulatory effect to account for the progressive changes in ventilatory components.