We have reported that minute ventilation [[Formula: see text]] and end-tidal CO(2) tension [[Formula: see text]] are determined by the interaction between central controller and peripheral plant properties. During exercise, the controller curve shifts upward with unchanged central chemoreflex threshold to compensate for the plant curve shift accompanying increased metabolism. This effectively stabilizes [Formula: see text] within the normal range at the expense of exercise hyperpnea. In the present study, we investigated how endurance-trained athletes reduce this exercise hyperpnea. Nine exercise-trained and seven untrained healthy males were studied. To characterize the controller, we induced hypercapnia by changing the inspiratory CO(2) fraction with a background of hyperoxia and measured the linear [Formula: see text] relation [[Formula: see text]]. To characterize the plant, we instructed the subjects to alter [Formula: see text] voluntarily and measured the hyperbolic [Formula: see text] relation ([Formula: see text]). We characterized these relations both at rest and during light exercise. Regular exercise training did not affect the characteristics of either controller or plant at rest. Exercise stimulus increased the controller gain (S) both in untrained and trained subjects. On the other hand, the [Formula: see text]-intercept (B) during exercise was greater in trained than in untrained subjects, indicating that exercise-induced upward shift of the controller property was less in trained than in untrained subjects. The results suggest that the additive exercise drive to breathe was less in trained subjects, without necessarily a change in central chemoreflex threshold. The hyperbolic plant property shifted rightward and upward during exercise as predicted by increased metabolism, with little difference between two groups. The [Formula: see text] during exercise in trained subjects was 21% lower than that in untrained subjects (P < 0.01). These results indicate that an adaptation of the controller, but not that of plant, contributes to the attenuation of exercise hyperpnea at an iso-metabolic rate in trained subjects. However, whether training induces changes in neural drive originating from the central nervous system, afferents from the working limbs, or afferents from the heart, which is additive to the chemoreflex drive to breathe, cannot be determined from these results.