Acutely ill patients commonly increase minute ventilation (V) to varying degrees. The pattern of breathing utilized to increase V was analyzed in normal subjects and acutely ill surgical patients. V = tidal volume (VT) x frequency (f), or V = inspiratory flow x (TI/TTOT, inspiratory time/total cycle time). CO2 inhalation and exercise were used to induce supine hyperventilation in normal subjects. This was compared to hyperventilation in acutely ill surgical patients. Measurements were made of O2 consumption, CO2 production, V, VT, f, inspiratory flow, TI, and TTOT. With small increases in V (up to twice control), normal subjects increased inspiratory flow and TI/TTOT with both CO2 and exercise. CO2 inhalation increased VT with no change in f, while exercise increased both VT and f. When V increased beyond twice control, TI/TTOT remained constant and increases in inspiratory flow accounted for the entire increase in V. In acutely ill patients with increased V, average f was elevated and VT decreased, but there was no constant relationship of f and VT with increasing V. However, TI/TTOT was relatively constant at 0.40- 0.46; therefore, in order to increase V, inspiratory flow had to increase! Patients also showed a tendency to breathe at a relatively fixed VT while normal subjects did not show this phenomenon, even with increases of up to 3 times control V. Continuous, rapid analysis of gas exchange and breathing patterns holds promise for early detection of patients with V inappropriate to metabolic demands, and serves as a sensitive indicator of abnormal patterns used by acutely ill patients to increase V.