Pulmonary blood vessels with diameters of 200∼400 μm produce considerably more force in response to vasoconstrictor drugs than those which are either smaller or larger. We have therefore investigated whether or not hypoxic pulmonary vasoconstriction (HPV) is more powerful in vessels of these diameters. We have also looked at the possibility that vessels from different regions of the lung respond differently. To do this we have grouped vessels according to their location within the lung as well as by size. We used a small vessel myograph (Cambustion AM10, Cambridge, UK) to study 208 preconstricted (1 μM PGF2α) small pulmonary arteries (300∼800 μm diameter when stretched to a tension equivalent to 25 mmHg transmural pressure) from 39 rats anaesthetized with 2% inspired halothane. A biphasic contraction was observed in response to hypoxia (ca. 25 mmHg Po2). The magnitudes of both the first, transient, phase (PT, peak tension) and of the second, sustained, phase (SST, steady state tension) were measured. The latter was measured 40 min after the start of hypoxia. The first phase was most pronounced in vessels with an average diameter of 423 μm while the second phase was most pronounced in larger vessels (mean diameter 505 μm). These maximal responses were all seen in vessels somewhat larger than reported by others. The responses of smaller vessels (400∼500 μm) did not depend upon their location within the lung, but those of larger vessels (600∼700 μm) showed regional differences. Those from the right lobe and those from the base of the lung gave the largest responses. It was especially noticeable that large vessels (631 μm diameter) from the base of the right lung gave the biggest responses. Thus HPV seems to occur not in a uniform manner, dependent solely to the size of vessels, but it also depends to some degree on the region of the lung from which vessels have been taken. Furthermore, our results suggest that larger vessels, as well as smaller ones, may contribute significantly to HPV.