Under conditions of varying flow rates, total myocardial blood flow, measured by fractional uptake of rubidium-84, using a coincidence counting system, was compared with myocardial flow measured by microspheres (15 +/- 5 micrometer). The methods were compared, open-chested, in 47 dogs: 17 during control, ten following 5 min of ligation of left anterior descending coronary artery, five following i.v. isoproterenol, six following ligation and isoproterenol, and nine after ligation plus dipyridamole. Regional flows by Rb-84 and by either Ce-141 or Cr-51 microspheres were also compared for left ventricle, as well as for nonischemic posterior wall, which served as a reference area, and for anterior wall with ligation of left anterior descending artery in the same preparations. There were no significant differences in total or regional flow measured by the two methods, nor in the estimate of ischemic area size. The data indicate that measurement of myocardial blood flow by fractional uptake of a potassium analog is a reliable method in the presence of ischemia and drug intervention. It is suggested that the inequalities of extraction ratio that occur with differing flow rates do not invalidate fractional-uptake methods over the flow ranges examined.