A double-antibody radioimmunoassay was developed for quantification of rat arginine-rich apolipoprotein in sodium decyl sulfate. Arginine-rich protein, labeled with 125I by the chloramine-T method, had the same chromatographic characteristics on Sephadex G-200 as unlabeled arginine-rich protein and up to 70% of 125I-labeled arginine-rich protein was precipitated by antisera to arginine-rich protein in rabbits. The assay is sensitive at the level of 1-10 ng and has intraassay and interassay coefficients of variation of 5.4 and 6.8%, respectively. The specificity of the assay was established by competitive displacement of 125I-labeled arginine-rich protein from its antiserum by arginine-rich protein and lipoproteins containing this protein, but not by rat albumin or other purified apolipoproteins. Immunoreactivity of rat serum and lipoproteins was complete as demonstrated by comparison with their delipidated form. The accuracy of the immunoassay was further substantiated by comparison with the amount of arginine-rich protein in chromatographic fractions of total apoprotein of very low and high density lipoproteins, and by recovery experiments in ultracentrifugally separated fractions of serum. In contrast to an immunoassay reported previously for rat apo A-I, sodium decyl sulfate was not required for complete immunoreactivity of serum and lipoproteins. The inclusion of sodium decyl sulfate (9 mM final concentration) was necessary, however, for stability of labeled and unlabeled preparations of arginine-rich protein. Content (weight %, means values +/- S.D.), of immunoassayable arginine-rich protein in isolated lipoproteins was 15 +/- 1.5% in very density lipoproteins; 6.8% in low density lipoproteins (1.02 less than d less than 1.04 g/m); 7.1 +/- 0.3% in high density lipoproteins; and 4.8 +/- 0.5% in lymph chylomicrons. Concentration in whole serum was 18.1 +/- 1.4 and 20.4 +/- 2.3 mg/dl for male and female rats, respectively. Only about 55% of arginine-rich protein was recovered in the major lipoprotein classes and about 40% was in "lipoprotein-free" serum (d greater than 1.25 g/ml). Among the lipoproteins, the high density lipoprotein fraction contained twice the amount of arginine-rich protein recovered in very low or low density lipoproteins (26.6 vs. 13.5 and 13.4%, respectively). The significance of the large amount of arginine-rich protein in the 1.25 g/ml infranatant fraction is not apparent. Although repetitive centrifugation did not alter the amount recovered in this fraction, the possibility of an artifact induced by centrifugation and high salt concentration cannot be excluded.