The evaluation of regional myocardial blood flow (RMBF) during cardiac catheterization is of particular diagnostic interest. The purpose of this investigation was to validate x-ray densitometric parameters for the evaluation of RMBF. In five anesthetized dogs, arterial flow in the circumflex coronary artery was measured continuously with an electromagnetic flowmeter, and RMBF was determined by colored microspheres. Five different perfusion levels were created by mechanical obstruction of the coronary artery or by intravenous infusion of adenosine. At each steady-state perfusion level, digital subtraction coronary angiograms were obtained for densitometric analysis. Results documented a close correlation between the related time parameters 1/Mean Transit Time (1/MTT, r2 = 0.969), and 1/Rise Time (1/RT, r2 = 0.965) and RMBF over a wide range between 0.36 ml/(min x g) and 11.16 ml/(min x g). Maximum myocardial contrast density (Imax) also showed a good, but inverse correlation (r2 = 0.889) with RMBF and, therefore, did not reflect vascular volume. Contrast medium Appearance Time (AT) showed no correlation to RMBF (r2 = 0.017). Repeat densitometric measurements for different perfusion levels revealed a good reproducibility for MTT (accuracy: 0.001 s; precision: 0.447 s or 6.7 %) and RT (accuracy: 0.014 s; precision: 0.202 s or 10.4 %), while AT (accuracy: 0.072 s; precision: 0.420 s or 68.5%) and Imax (accuracy: 0.022 GL; precision: 1.197 GL or 44.5%) showed substantial variation. Myocardial perfusion reserve (MPR) calculated from RT (r2 = 0.90) or MTT (r2 = 0.94) showed better correlations to RMBF reserve than MPR calculated from AT (r2 = 0.04). In conclusion, only 1/MTT and 1/RT showed a good reproducibility and a close correlation to RMBF. Therefore, only these parameters can be recommended for calculations of RMBF and its reserve under clinical conditions.