Self-superantigens have been described as products of endogenous retroviruses of the mouse ('minor lymphocyte stimulating loci') that are capable of interacting without prior processing with conserved domains of TCR V beta chains, causing the activation and deletion of most T cells expressing products of determined V beta gene families [1-4]. The fact that superantigens activate a far higher percentage of T cells (1-20%) than conventional, peptidic antigens (< 0.1%) provides the methodological advantage that the degree of clonal deletion may be measured by the analysis of the TCR repertoire using appropriate anti-V beta antibodies. Although much information on the spatio-temporal organization of repertoire-purging has been gathered by virtue of self-superantigens, serious doubts exist as to the possibility that such structures serve as pathogenetically relevant autoantigens. Thus, certain inbred mice spontaneously develop autoimmune diseases, although they bear T-cell repertoires that appear to be purged from self-superantigen-reactive V beta products. In addition, therapeutic interventions targeted to V beta gene products that are not specific for self-superantigens are successful in preventing disease development. The lack of correlation between superantigen-related V beta deletions and autoimmune disease development is substantiated in further models of murine autoimmunity. Based on these observations, we formulate the hypothesis that self-superantigen-reactive T cells are not involved in the development of autoimmune diseases.