A rabbit antiidiotypic antiserum raised against an A.SW IgG1K monoclonal anti-sperm whale myoglobin (Mb) antibody, HAL19, and extensively absorbed with normal mouse immunoglobulin and MOPC 21 (IgG1K), was found to detect a common or major anti-Mb idiotype expressed by some but not all anti-Mb monoclonal antibodies, regardless of immunoglobulin G (IgG) subclass, and by 40-50% of the anti-Mb antibodies in immune serum from five high responder strains of mice representing five different Igh allotypes. It did not inhibit antibodies to three unrelated protein antigens. The fraction of antibodies expressing this idiotype, denoted IdHAL19, was regulated by H-2-linked genes that correlated exactly in four independent haplotypes and an F1 with the known Mb immune response (Ir) genes and may be identical to these. Whereas less than 50% of antibodies from high responder mice were inhibitable by anti-IdHAL19, greater than 80% of antibodies from low responder mice, tested at comparable final antibody concentration, were inhibitable. This result was true for both low responder haplotypes, H-2b (B10) and H-2k (B10.BR). The idiotype was found to be present on antibodies that bound to native Mb but not fragments 1-55 or 132-153 of Mb or a denatured form, S-methyl Mb. This specificity for native Mb paralleled that of the monoclonal idiotype HAL19 itself. Therefore, the production of antibodies specific for native in contrast to denatured Mb was studied in H-2-congenic high and low responder strains. Strikingly, low responders produced antibodies that reacted almost exclusively with the native conformation, whereas a larger proportion of antibodies from high responder mice also reacted with the denatured form, S-methyl Mb. Bypassing of the Ir gene defect by immunization with Mb attached to a carrier, F gamma G, resulted in low responder antisera resembling higher responder sera in both idiotype expression and conformational specificity. The simplest explanation of these results is that H-2-linked Ir genes control antibody fine specificity, which is reflected in the idiotypes of the variable regions expressed. We suggest that low responder mice produce a more limited repertoire of antibodies consisting primarily of IdHAL19-positive antibodies specific for the native conformation of Mb. High responder mice produce a greater diversity of antibodies to Mb, so that the IdHAL19-positive, conformation-specific population represents a smaller proportion of the total. Similarly, the use of carrier-specific helper T cells in low responder mice results in a greater diversity of antibodies, which dilutes out the IdHAL19 subset.(ABSTRACT TRUNCATED AT 400 WORDS)