Platelet-associated IgG (PAIgG) was studied by a quantitative platelet radioactive anti-IgG test (PRAT) in 298 patients. At the time of investigation, 171 patients were thrombocytopenic (platelet count < 100 X 10(9)/l), 127 had normal platelet counts. Patients fell into the following disease categories: Idiopathic thrombocytopenic purpura (ITP) (N = 81), possible ITP (19), acute ITP (9), systemic lupus erythematosus (22), autoimmune haemolytic anaemia of warm-type (18), systemic blood disease (65), liver diseases (35), other (49). A significant elevation of PAIgG was found in all disease categories. There was a significant correlation between PAIgG and the reciprocal values of platelet counts for most disease groups. No relationship was discernible between PAIgG and hypergammaglobulinaemic states (serum IgG > 1.8 g/l). Platelet survival studies (N = 30) revealed that normal and increased values of PAIgG were associated with normal or shortened platelet mean life span. It is concluded that an elevated PAIgG is only one of several factors involved in the development of immunologically medicated thrombocytopenia.