The efficacy and toxicity of oral azidothymidine has been studied in 145 patients with the acquired immunodeficiency syndrome (AIDS) or AIDS-related complex (ARC). The median survival time of AIDS patients on azidothymidine was 4.5 times higher when compared to a historical AIDS group who had not received the drug. This result must be interpreted with caution because of changes in treatment of HIV infection and growing awareness of AIDS which may have led to earlier diagnosis in the group treated with azidothymidine. The mortality was significantly higher in those patients who received transfusions and was particularly high in those who were transfused before azidothymidine. There was a significant difference in the occurrence of opportunistic infections in the patients who received transfusions compared with those who did not. AIDS patients treated immediately after an episode of Pneumocystis carinii pneumonia survived significantly longer than those in whom treatment was delayed for three months or more, and longer than those who were treated with azidothymidine because of another opportunistic infection or Kaposi's sarcoma. The T4 cell median counts increased in patients treated with azidothymidine reaching a peak at the end of the fourth month of treatment in the ARC group and at the end of the first month in the AIDS group with a subsequent fall in both groups. Sixty per cent of patients were p24 viral antigen positive at the start of treatment and 19 per cent of these patients had a fall of more than 50 per cent in antigen level while 32 per cent became antigen negative following treatment with azidothymidine. The mortality in the patients where the antigen disappeared or in whom there was a major fall of more than 50 per cent in antigen level was significantly less than in those where there was no change in antigen level. Twenty-nine patients were treated with azidothymidine because of skin Kaposi's sarcoma and in 17 tumour regressed or was stable. Thirty-two per cent of patients treated with azidothymidine became anaemic. Neutropenia occurred in 3 per cent of patients. Platelets increased initially after treatment but subsequently fell to thrombocytopenic levels in eight patients. Nine of 12 patients with thrombocytopenia before azidothymidine was commenced responded with an increased platelet count.