Many lines of evidence suggest that autoimmune diseases result from chronic immune activation following environmental exposures in genetically susceptible individuals. A genetic basis for autoimmunity is supported by twin and family studies, candidate gene investigations, animal models, and whole genome microsatellite scans. These findings predict, and clinical observations support, familial clustering of a number of individual autoimmune diseases, notably lupus, multiple sclerosis, type-1 diabetes mellitus, rheumatoid arthritis, and recently the idiopathic inflammatory myopathies. Yet, not only is the same autoimmune disease increased in prevalence in pedigrees of persons affected with a given disorder, but other autoimmune diseases are as well. We review these data and propose a hypothesis consistent with these findings. This model posits that a rheumatic disease, as currently classified, is actually composed of a number of elemental disorders. Each of these is defined by the minimal necessary and sufficient environmental exposures and genes that result in a pathology leading to a given sign-symptom complex.