Anti-Sm and anti-ribosomal P protein antibodies show a high degree of specificity for the disease SLE. To determine whether a relationship between these two autoantibodies existed, the frequency of anti-P was determined in sera with and without anti-Sm activity. Of sera from lupus patients with anti-Sm 18/65 (28%), and 6/55 (11%) of sera without anti-Sm had anti-P as determined by an ELISA using a recombinant P2-beta-galactosidase fusion protein as Ag (p less than 0.05). The levels of anti-P were significantly higher in sera containing anti-Sm (0.37 +/- 0.45) than in sera without anti-Sm antibodies (0.18 +/- 0.20) (p less than 0.01). Similarly, a significantly higher proportion of anti-P positivity was found in autoimmune MRL/Mp-lpr/lpr mice positive for anti-Sm (11/53 = 21%) compared to age- and sex-matched mice without anti-Sm (3/53 = 6%) (p less than 0.05). The IgG subclass distributions for anti-Sm and anti-P antibodies were similar in the MRL mice (IgG2a greater than IgG2b greater than IgG3 greater than IgG1). The association did not reflect polyclonal B cell activation in a proportion of MRL mice because no significant differences were observed in anti-DNA, antichromatin or total serum IgG levels in mice with and without anti-Sm or, in mice positive for both anti-P and anti-Sm compared to mice positive for anti-Sm alone. Cross-inhibition experiments excluded the possibility that the Sm and P protein Ag shared a common epitope. Longitudinal measurement of anti-P and anti-Sm antibody levels by ELISA in three mice indicated that both antibodies first appeared at about 3 to 4 mo of age and fluctuated two- to threefold over 3 to 8 mo with independent peaks of activity. Recent observations regarding a relationship between anti-Sm and autoantibodies to other ribosomal proteins suggest that the association may be explained by an immune response to epitopes coassociated on the ribosome.