There is now little doubt that functional immune polymorphism exists and can exert a significant effect on the severity of immunologic disease. Studies in this area are driven by the belief that an understanding of the influence of immune polymorphism on immunologic disease will improve our ability to predict disease occurrence and severity. The usefulness of this information is necessarily limited, however, by the fact that an individual has a fixed genome. If an individual has an allele which predisposes to increased disease severity, there is little we can do, other than treat the disease in that individual in a more aggressive manner or attempt to find preventative therapies. There is, however, one unique immunologic disorder that is an exception to this rule. Graft-versus-host disease (GVHD), the major complication of bone marrow transplantation, is an immunologic reaction mediated, in large part, by donor T cells. It is unique in that it is the only immunologic disorder in which we can choose the genetics of the immune system causing the disorder. In this disease, more than any other, an understanding of the role of immune polymorphism is essential since we could hypothetically use this information to choose a donor whose T cells will not mediate GVHD of life-threatening severity. In this review, I argue that there is highly suggestive evidence that allelic polymorphism in non-MHC genes encoding mediators of the immune response exerts a very significant effect on the severity of GVHD and that typing for such allelic polymorphisms will in future be as important as MHC-typing in choosing an appropriate donor for bone marrow transplantation. Genes and Immunity (2000) 1, 316-320.