To confirm the effects of physical training and detraining on CO2 chemosensitivity, we followed hypercapnic ventilatory response at rest in the same five subjects during pre-, post- and detraining for 6 years. They joined our university badminton teams as freshmen and participated regularly in their team's training for about 3 h a day, three times a week, for 4 years. After that they retired from their teams and stopped training in order to study in the graduate school for 2 years. Maximum pulmonary ventilation (VEmax) and maximal oxygen uptake (VO2max) for each subject were determined during maximal treadmill exercise. The slope (S) of ventilatory response to carbon dioxide at rest was measured by Read's rebreathing method. Mean values of VEmax increased statistically during training and decreased statistically during detraining. A similar tendency was observed in VO2max. The average value of S before training was 1.91 l.min-1.mmHg-1, (+/- ) SD 0.52 and it decreased gradually with increasing training periods; the difference between the S values before (1980) and after training (1982, 1983 and 1984) were all significant. Furthermore, the mean values of S increased significantly during detraining as compared with those obtained at the end of training (April 1984). We concluded that in normal subjects, long-term physical training increases aerobic work capacity and decreases CO2 ventilatory responsiveness, and that the ventilatory adaptations with training observed here are reversible through detraining.