The effect of presensitization was studied in a consecutive series of 119 first cadaver renal transplants. Actuarial graft survival for the entire series was 76% at 1 year, 71% at 2 years, and 70% at 5 years after transplantation. Presensitized patients fared less well (53% graft survival for 2–5 years) than apparently nonsensitized patients (83 and 79% graft survival for 2 and 4 years, respectively). Analysis of subgroups of sensitized patients did not reveal a simple parallel relationship between the degree of presensitization (as expressed by the percentage of a random donor panel of lymphocytes reacting with the patients' pretransplant sera) and the outcome of the graft. With one exception, all patients with antibodies to >70% of the panel lost their grafts. However, the subgroup with antibodies to an intermediate percentage (29–48%) of the panel had a favourable course (76% had 5-year graft survival), only slightly inferior to that of nonsensitized patients, and better than that of patients reacting to < 25% of a panel. Although graft loss in presensitized patients appeared to be confined to patients with mismatch-specific antigraft antibodies, a substantial number of such patients, and also of patients experiencing the accelerated rejection syndrome (in whom antigraft antibodies were operationally significant) fared very well after transplantation. Identification of the mechanism involved, and the ability to predict the outcome in the individual case, would not only be of considerable theoretical interest, but could have a marked practical effect on any clinical transplantation programme.