Overall open-loop gain of the CO2-ventilation feedback control system in hypoxia (GHco2) was determined on 8 male and one female healthy subjects. They breathed in a closed circuit, and were subjected to the progressive hypoxia test. This procedure was first conducted without dead space (DS), then with 250, 500, and finally 750 ml DS, consecutively. GHC02 was calculated by dividing the slope of the CO2 response curve (S) by that of the metabolic hyperbola (SL). GHco2 was considerably larger than the overall open-loop gain of the 02-ventilation feedback control system (Go2) previously obtained. This was ascribed to the facts that S was larger than the slope of the hypoxia response curve, and that the absolute value of SL in GHco 2 was smaller than that in G02. SEVERINGHAUS (1972) was the first to present the concept of the overall gain of the C02-feedback control system. The same idea was also proposed as the open-loop gain by MILHORN (1966). The concept of overall gain corresponds to that of open-loop gain. In the present study, we defined the overall open-loop gain as GHco2 (defined as overall gain in our previous study) by dividing the slope of the CO2-ventilation response line (S) by the slope of the metabolic hyperbola (SL) at a given end-tidal Pco2 (PETco2). When a disturbance is introduced into the system in order to change PETco2 by 4PETco2, the final change in PETC02 will be reduced to 4PETc o 2/(l + GHc o 2).In the previous communication (MASUYAMA et al., 1983), overall open-loop gain of the 02-ventilation feedback control system was evaluated and compared with that of the C02-ventilation feedback control system in normoxia. The latter was found to exceed the former, suggesting the predominant role of CO2 in the control of ventilation. However, it is known that CO2 sensitivity in regulation of respiration is affected by the presence of hypoxia (NIELSEN and SMITH, 1951).To assess the actual significance in the feedback control system in hypoxia, both