SUMMARY. When steady state pressure-flow relations are studied in the circumflex coronary artery, pressure gradients develop between it and other branches of the left coronary artery. To assess the effects of these pressure gradients, we compared the pressure axis intercept and shape of steady state circumflex pressure-flow relations in the presence and absence of gradients after autoregulation was abolished, both in the beating heart and during long diastoles in dogs. We used peripheral coronary pressures and radionuclide-labeled microspheres to assess arterial collateral flow. In the beating heart, interarterial pressure gradients reduced the curvature at low circumflex pressures, and overestimated the mean pressure axis intercept by 7.8 mm Hg (P < 0.05). The results were similar for the pressure-flow relations derived during long diastoles. This overesrimation exaggerates the difference between the pressure axis intercept and coronary sinus pressure. The peripheral coronary pressure and microsphere results indicate that these effects are mediated largely by arterial collateral flow. {Circ Res 56: 11-19, 1985) CORONARY arterial pressure-flow relations have been used to investigate the physical forces that regulate coronary blood flow (Cross et al., 1961;Downey and Kirk, 1975;Bellamy, 1978). The pressure axis intercept of these relations has been interpreted as the downstream pressure opposing forward flow in the coronary circulation; however, this pressure has greatly exceeded the coronary sinus pressure, which traditionally has been assumed to be the downstream pressure. The shape of these relations, found to be linear by some (Dole and Bishop, 1982;Eng et al., 1982) but curvilinear by others (L'Abbate et al., 1980;Klocke et al., 1981), is thought to convey information about the conductance of the coronary circulation. Because of this variability of the pressure axis intercept and the shape of coronary arterial pressure-flow relations, the physiological significance of both remains uncertain.One factor that might account for this variability is the specific coronary artery in which the pressureflow relation is obtained. In our laboratory, we have consistently found that the pressure axis intercept of a steady state circumflex pressure-flow relation (Verrier et al., 1980;Vlahakes et al., 1982) is much higher than that in a steady state left main coronary arterial pressure-flow relation (Rouleau et al., 1979;Uhlig et al., 1984). A possible explanation for the higher pressure axis intercept of the circumflex pressure-flow relation may be that when circumflex pressure is selectively lowered below that in the other branches of the left coronary artery, interarterial pressure gradients occur and are accompanied by collateral flow that is not measured by the upstream flow transducer. This error in measurement could underestimate total circumflex flow and thus overestimate the pressure axis intercept. Whether these pressure gradients also affect the shape of a circumflex pressure-flow relation is unknown.To assess...