Mammary carcinoma tissue from 514 primary breast cancer patients were all biochemically and histochemically analyzed for both estrogen receptors and progesterone receptors. The dextran-coated charcoal (DCC) method measured the ER and PR as defined by Scatchard analysis, ligand competition experiments and target organ specificity. The ligands, estradiol-6-carboxymethyloxime-BSA-fluoresceine isothiocyanate and hydroxyprogesteronehemisuccinate-BSA-tetramethylrhodamine isothiocyanate, used for histochemistry, did not bind to either ER or PR and were mainly bound to the membrane fraction of isolated breast cancer cells. Fluorescence was not specifically inhibited by estrogens or progestogens. In addition, "estrogenic" always coincided with "progestogenic" fluorescence. The binding of the fluoresceine compounds to tissue slides depended on the large steroid hormone substitution on the bovine serum albumin molecule. Clinical parameters, known to be related to ER and PR did not correlate with the histochemical results. The observations indicated the impossibility of specific steroid receptor detection by the histochemical method. Therefore, up to the present, evaluation of hormone dependency and prognosis in human breast cancer cannot be based on this approach.