After a century of research and despite intensive scrutiny, the origin of autoantibody production remains an enigma. Recently, the essential role of B cells in promoting systemic autoimmunity in mice seems more important than previously thought: self-reactive B cells can be subject to positive selection and a deficiency in serum IgM predisposes to the development of IgG antibodies to autoantigens. Studies of the B cell repertoire expressed in systemic autoimmune diseases have provided important clues. In human lupus, quantitation of this repertoire reveals the presence of an expansion of IgG clonotypes that impart reactivity with disease-related autoantigens. The nucleotide sequences of autoantibodies derived from these patients and expressing nephritogenic idiotopes (present in immune complexes and renal eluates of subjects with active disease) show features of diversification with a high rate of replacement/silent mutations and clustering of the mutations in the hypervariable regions, suggesting than an antigen-driven process plays a role in the generation of pathogenic autoantibodies. Currently, the contributions of apoptosis and of cell receptor signaling to this triggering are being appreciated. Pursuing these and related issues will have an important impact on autoimmune research.