In a randomized double blind study, the arterial changes produced either by hydrochlorothiazide plus amiloride (Group I), or by hydrochlorothiazide plus captopril (Group II) were investigated in two territories of the arterial tree, the common carotid artery and the terminal aorta. Arterial echo-tracking techniques of high resolution and applanation tonometry were used to evaluate non-invasively the indices of arterial stiffness and carotid wave reflections. In Group I and II, there was a similar significant decrease in brachial blood pressure (BP) and carotid diastolic diameter and an increase in aortic compliance and distensibility. Groups I and II differed significantly in aortic diastolic diameter which decreased in Group I but not in Group II, and in carotid wave reflections which were modified in Group II but not in Group I. Thus, captopril associated with hydrochlorothiazide resulted in a shift in the carotid arterial reflection wave from systole to diastole with no reduction in the aortic diastolic dimension. For similar BP reduction, the combination of hydrochlorothiazide and amiloride had no significant effect on the carotid reflection wave, but caused a significant reduction in the aortic diastolic diameter. These intergroup differences were related to the presence or absence of converting enzyme inhibition.