The methodology for the collection, extraction, separation and measurement of urinary angiotensin II [the octapeptide, ANG(1-8)] is described. To determine the origin of urinary ANG(1-8), mean arterial pressure, renal hemodynamics and the arterial, renal venous and urinary concentrations of ANG(1-8) were examined prior to and following the constant intra-arterial infusion of tritiated angiotensin II [3H-ANG(1-8)] in graded doses of 0.5,2.0 and 2.5 ng/kg/min in 5 uninephrectomized, anesthetized female dogs. The infusion of 3H-ANG-(1-8) had no significant effect on mean arterial pressure, glomerular filtration rate, renal blood flow or urine flow rate. The mean concentration of ANG(1-8) in the urine was 3.7 fmol/ml. None or only trace amounts of 3H-ANG(1-8) were detected in the urine in spite of marked increases in renal arterial 3H-ANG(1-8) concentrations. These observations suggest that urinary ANG(1-8) was derived de novo from the intrarenal generation of angiotensin II. In addition, plasma and urinary concentrations of ANG(l-8) were assessed in patients with essential hypertension undergoing treatment with either a diuretic (n = 14) or an angiotensin-converting enzyme inhibitor (n = 14). Although the concentrations of plasma ANG(1-8) responded appropriately to the respective therapies, the urinary excretion of ANG(1-8) was not different following either therapy. These data suggest that ANG(1-8) collected from the urinary bladder may not occur in adequate concentrations to accurately assess the activity of the intrarenal reninangiotensin system.