The thymus has been studied in myasthenia gravis patients to assess the effects of previous immunosuppression on total yields of cell suspension, immunohistology and culture responses. The reduction in cell yields by pretreatment with corticosteroid was very variable. In 16 of 32 cases, cortical, medullary and total cell numbers were all greatly reduced ('depleted cases'), whereas in the others, they were within or near the typical range for untreated myasthenics. Cortical thymocytes were even more depleted than precursor thymic blasts. Thus the interpatient differences in sensitivity to corticosteroid recently described for mature T cells also affected immature cortical thymocytes and their differentiating medullary progeny. In the medulla, mature (CD3+) T lymphocytes and germinal centres were enriched by the loss of cortex and appeared relatively healthy, but somewhat depopulated. Concomitantly, in-vitro T-cell responses to acetylcholine receptor (AChR) and production of anti-AChR antibody and total IgG by thymic cells were usually well within the typical range (assessed per 10(6) cells). Moreover, the total productivity of the entire thymus was reduced almost entirely by the cellular depopulation rather than by decreased function per surviving cell. Thus the main actions of this alternate day therapy with corticosteroids were apparently on total peripheral cell numbers, and perhaps on activated cells and effector mechanisms too, and its thymic effects were inessential.