Careful objective measurements of ratios between renal artery diameter at several points and renal length, width and the product of length and width were useful In the diagnosis of renovascular hypertension in 4 of 5 persons cured or greatly improved by operation, but use of these ratios led to incorrect diagnosis in 3 out of 4 patients with renal artery lesions and with negative conventional split function and pyelogramurea washout tests who had no other evidence of renovascular hypertension. While greater sophistication and perhaps more numerous measurements and comparisons might lead to Increased predictive accuracy, it is felt that 1 or several of the available tests designed to show the characteristic altered renal physiology should be employed in addition to aortography in the selection of patients for surgery.