With a pulsed Doppler imaging system, it is now possible to interrogate sites from the aorta to the popliteal trifurcation. To determine which velocity parameters could be correlated with the degree of disease as determined by angiography, 34 arterial stenoses identified by scanning were also evaluated by contrast arteriography and classified in 10% increments. The angiographic readings were blinded with respect to the scan results. Four hand-measured parameters from velocity waveforms obtained at the site of stenosis were correlated with the angiogram--peak systolic velocity, systolic rise time, diastolic reverse velocity, and diastolic reverse flow time. When diastolic reverse flow was absent, diastolic forward flow was recorded. To describe diastolic flow along a continuum, diastolic reverse velocity was ascribed a positive value and diastolic forward velocity was ascribed a negative value. A systolic velocity gradient (peak velocity/rise time) was also calculated. The relationship between the angiographic categories and the measured parameters was evaluated with the Jonkheere-Terpstra trend test. A trend was determined with diastolic flow (diastolic reverse flow or diastolic flow velocity) that was significant (p less than 0.01). The linear regression was calculated (y = 40.8 + [-5.6X]), and correlation coefficient was obtained (r = 0.76) that was statistically significant (p less than 0.01). The method enables mapping and calculation of arterial stenoses by noninvasive means. This can be expected to obviate the need for diagnostic angiograms in certain and select cases in which angioplasty can be expected to be beneficial. It also affords a convenient quantitative means of following lesions over time.