We investigated the relationship between minute ventilation (VE) and net respiratory muscle pressure (Pmus) throughout the breathing cycle [Total Pmus = mean Pmus, I (inspiratory) + mean Pmus, E (expiratory)] in six normal subjects performing constant-work heavy exercise (CWHE, at approximately 80% maximum) to exhaustion on a cycle ergometer. Pmus was calculated as the sum of chest wall pressure (elastic + resistive) and pleural pressure, and all mean Pmus variables were averaged over the total breath duration. Pmus, I was also expressed as a fraction of volume-matched, flow-corrected dynamic capacity of the inspiratory muscles (P(cap, I)). VE increased significantly from 3 min to the end of CWHE and was the result of a significantly linear increase in Total Pmus (Delta = 43 +/- 9% from 3 min to end exercise, P < 0.005) in all subjects (r = 0. 81-0.99). Although mean Pmus, I during inspiratory flow increased significantly (Delta = 35 +/- 10%), postinspiratory Pmus, I fell (Delta = -54 +/- 10%) and postexpiratory expiratory activity was negligible or absent throughout CWHE. There was a greater increase in mean Pmus, E (Delta = 168 +/- 48%), which served to increase VE throughout CWHE. In five of six subjects, there were significant linear relationships between VE and mean Pmus, I (r = 0.50-0.97) and mean Pmus, E (r = 0.82-0.93) during CWHE. The subjects generated a wide range of Pmus, I/P(cap, I) values (25-80%), and mean Pmus, I/P(cap, I) increased significantly (Delta = 42 +/- 16%) and in a linear fashion (r = 0.69-0.99) with VE throughout CWHE. The progressive increase in VE during CWHE is due to 1) a linear increase in Total Pmus, 2) a linear increase in inspiratory muscle load, and 3) a progressive fall in postinspiratory inspiratory activity. We conclude that the relationship between respiratory muscle pressure and VE during exercise is linear and not curvilinear.