The effect of phlebotomy on 44 patients (33 men and 11 women) with porphyria cutanea tarda (PCT) is described. All had typical biochemical and clinical signs of PCT. The patients were bled until slight anemia and signs of iron deficiency. In patients who were bled frequently and who were not given iron supplements in conjunction with phlebotomy, urinary uroporphyrin excretion decreased to 1 mg (or 1.2 mumol) a day or less in 1-8 months (mean 4 months) and then continued to decrease long after the treatment had been stopped and became normal in many patients. A decrease in porphyrin excretion comparable to that in patients treated with phlebotomy occurred in only two out of 12 controls. Remission occurred in patients treated with phlebotomy who had iron overload as well as in patients with quantitatively normal iron stores and in patients abstaining from alcohol as well as in patients who continued their abuse of alcohol. The disappearance of skin fragility and ulcers coincided roughly with the time when the urinary excretion of uroporphyrin decreased to about 1 mg (1.2 mumol) a day. After 43 remissions (in 43 patients) not followed by iron medication, one relapse occurred after a period of 20 months. In 5 patients in whom their iron stores were replenished a relapse occurred in all within 18 months. After longer observation periods relapses occurred also after remissions not followed by iron medication. During observation periods of 3-10 years 15 out of 41 patients relapsed biochemically. These had re-accumulated iron stores spontaneously. The results indicate that phlebotomy is consistently effective and probably exerts its effect mainly by reduction of iron stores.