It is generally assumed that the immunological activity of the follicle-stimulating hormone (FSH) molecule, as reported in the radioimmunoassay (RIA), is identical to the molecule's biological activity. To test the validity of this assumption, the receptor-binding and immunological activities of FSH in the serum of 35 infertile males and 11 fertile males were determined by radioreceptor assay (RRA) and by the standard RIA. The results were analyzed in terms of the ratio of binding (RRA) to immunological (RIA) activity x 100%, referred to as the B/I%, for each patient based on data from the RRA and RIA, respectively. The B/I% for fertile men ranged from 44% to 113% (mean of 80%). In the group of 11 infertile men with normal FSH levels by RIA (50 to 300 ng/mL), there was no significant difference in B/I% (range of 27% to 99%, mean of 59%) from the fertile controls. However, a statistically significant decrease in B/I% (range of 7% to 35%, mean of 18%) was noted in the 24 hypergonadotropic infertile men with RIA levels of serum FSH greater than 300 ng/mL. These significant discrepancies between receptor-binding and immunological activities of serum FSH raise the question of whether RIAs alone are valid parameters for the endocrine evaluation of infertile men with elevated serum FSH levels.