In this study we examined the preoperative value of the clonidine-suppression test in 15 patients with surgically proved pheochromocytomas. The result of the clonidine-suppression test was pathological (epinephrine plus norepinephrine above 500 ng/l 3 h after clonidine) in 10 of 15 patients (66%). These patients had relatively large tumors and higher basal norepinephrine plasma levels. Out of the 5 cases without a pathological clonidine test 4 had normal basal plasma catecholamine levels with the result that the clonidine test could not be properly applied and 1 case produced a false negative result. These 5 cases generally had smaller tumors and lower plasma catecholamine levels. Two of these cases had basally raised epinephrine values. The other three cases had either a paradoxical increase or a suspiciously low fall (less than 25%) in norepinephrine within the normal range. We conclude that the clonidine-suppression test is only reliable for the diagnosis of relatively large pheochromocytomas.