The author tested the hypothesis that in a canine lobar hypoxic pulmonary vasoconstriction (HPV) model, the passage of time alone would eliminate a previously observed association between increasing lobar HPV and repeated intermittent hypoxia of the lung lobe. The HPV model included electromagnetic measurement of the fraction of the cardiac output perfusing the left lower lobe (QLLL/Qt) and ventilation of the left lower lobe (LLL) independent of, but still synchronous with, the rest of the lung. Following surgical preparation of the model, the LLL and the rest of the lung were ventilated with 100% O2 and no further manipulations or procedures were performed for 120-150 min. The LLL then was made intermittently hypoxic four times by ventilation with 95% N2 and 5% CO2 and the LLL HPV response was quantified as the per cent decrease in QLLL/Qt. The LLL was kept either normoxic or hypoxic until the QLLL/Qt ratio was stable for several minutes. The first three LLL hypoxic exposures caused a significant progressive increase in LLL HPV response (from 37.8 to 54.7 to 61.3%) while the second LLL HPV response required significantly less time (16.6 min) to reach a stable decreased QLLL/Qt value compared with the first LLL HPV response (26.4 min). Animals with the smallest initial LLL HPV response increased their HPV response the most, and animals with the largest initial LLL HPV response increased their HPV response the least with repeated LLL hypoxic exposures. The conclusion that intermittent hypoxia increases HPV has important implications for the conduct of HPV experiments and the interpretation of blood-gas changes during one-lung ventilation for thoracic surgery.