Mechanisms involved in the control of pulmonary ventilation were studied in seven male subjects following 6 min of exercise on a cycle ergometer at 98w. Circulation to the legs was occluded by thigh cuffs (27 kPa) during the last 15 s of exercise and the subsequent 4 min of recovery. Respiratory gas exchange and the tidal partial pressures of O2 and CO2 were measured breath-by-breath. The results were compared to control studies without occlusion. There was a significant increase in both systolic and diastolic blood pressures during occluded recovery. Following occlusion systolic pressure remained elevated while diastolic pressure returned to control values. Occlusion during recovery caused hyperventilation during the first 1.5 min after exercise as evidenced by significantly higher VE/VCO2, VE/VO2, PETO2, and lower PETCO2. Following the release of the cuffs PETCO2, VE, VCO2, VO2, and heart rate all increased significantly above control values, while PETO2 decreased. PETCO2 rose abruptly 14.5 +/- 0.9 s after the release of the cuffs. Marked increases in VE and heart rate were seen, and occurred 30.8 +/- 1.5 s and 12.8 +/- 1.3 s, respectively, after cuff release. The 16.3 +/- 1.4 s lag between the increase in PETCO2 and VE after occlusion suggests that the ventilatory response to a sudden load of hypercapnic blood is not mediated by a pulmonary chemoreceptor. Other receptors, probably the peripheral chemoreceptors, appear to be responsible for hypercapnic hyperventilation.