The investigation of vascular impotence requires a simple, reliable and non-invasive method of estimating penile arterial supply. A method for measuring flow to the dependent portion of the penis has been developed using technetium-labelled red blood cells (99mTc-RBC). Its use, in conjunction with the intracavernous injection of papaverine, was investigated in 32 patients with various causes of impotence. Intracavernous papaverine was administered and a pneumatic cuff was inflated around the base of the penis both to isolate the penile circulation and to prevent dispersion of papaverine. The remaining blood pool was labelled with 99mTc-RBC. A gamma camera was used to produce a time-activity curve for the penis following release of the cuff. Penile blood flow was calculated from the initial gradient of the time-activity curve. Papaverine-stimulated penile blood flow was significantly greater in patients without evidence of arterial insufficiency. None of the subjects with arterial disease achieved flows greater than 20 ml/min/100 ml, while flow in patients without arterial disease exceeded this value. Flow estimation without the use of papaverine was not able reliably to discriminate between the 2 groups. There was no difference in papaverine-stimulated blood flow between impotent subjects with veno-sinusoidal leakage, in the absence of arterial disease, and a control group who achieved full erection with papaverine. It was therefore possible, by assessing the response to papaverine and measuring arterial blood flow, to distinguish between arterial insufficiency, veno-sinusoidal leakage and non-vascular causes of impotence. The method requires little skill and can be modified for use with inexpensive equipment.