To determine whether intracavernous injection of papaverine can discriminate vascular versus psychogenic impotence 80 mg. papaverine were injected intracorporeally into 27 impotent patients without any hormonal or neurological abnormality. The patients also underwent dynamic cavernosography, the artificial erection test and selective internal iliac arteriography. In addition, 21 of the patients underwent nocturnal penile tumescence monitoring. The response to intracavernous papaverine injection was noted as positive (fully rigid erection), intermediate or negative (soft or absent erection). Among 14 patients with severe arterial lesions and/or severe venous leakage the response was negative in 11 and positive in none. Among the remaining 13 patients (moderate or absent vascular abnormalities) the response was positive in 4 but negative in 4. Of 15 patients with an abnormal nocturnal penile tumescence test the response to intracavernous papaverine injection was negative in 12 and positive in none. However, of 6 patients with a normal nocturnal penile tumescence test the response was positive in only 1. Over-all, the results of the intracavernous papaverine injection test are correlated only fairly with those of the nocturnal penile tumescence test and the vascular investigations. This test could be useful to save some nocturnal penile tumescence tests in cases of suspected vascular impotence. However, the safety of this procedure must be confirmed in larger series. Indeed, 1 of our patients suffered priapism after a second intracavernous papaverine injection performed for a therapeutic purpose.