Androgen binding has been studied in the quadriceps femoris of recently castrated adult and intact immature male and female rats using a variety of techniques for separating and measuring hormone-receptor complexes. [3H]Testosterone, [3H]androstanolone (or 5 alpha-dihydrotestosterone). [3H]methyltrienolone (a potent synthetic androgen), and [3H]estradiol bind to the androgen receptor. Affinities are identical for the first two hormones (Kd = approximately 70 pM) and lower for estradiol (Kd = approximately 0.2 nM), as determined by Scatchard plots of binding data. Competition experiments indicate that in addition to the nonradioactive steroids corresponding to the above-cited tritiated compounds, progesterone, cyproterone acetate (an antiandrogen), and spironolactone compete for [3H]androgen binding by the receptor, but diethylstilbestrol, moxestrol (a potent synthetic steroidal estrogen), and cortisol do not. 3 alpha- and 3 beta-androstanediols slightly inhibit testosterone binding. Therefore, striated muscle androgen receptor specificity is identical to that of all androgen receptors of target tissues which have been previously studied. Binding is abolished by pronase and heat treatment, and displays an approximate 7S sedimentation coefficient in low salt ultracentrifugation gradient analysis. Preliminary observations suggest hormone-induced receptor translocation into the nucleus. No evidence has been found for an independent estrogen receptor. In the course of the binding experiments, extensive metabolism of androstanoloe and testosterone was observed in muscle cytosol at 0-4 C, during the 2-h incubation period used for most binding studies. Metabolite formation can jeopardize the binding data, specifically altering the significance of competition experiments with relatively high concentrations of steroids approaching the Km of metabolizing enzymes. Therefore, most quantitative studies were performed in enzyme-free, receptor-containing cytosol preparations. In adult male rats castrated for 2 days, the concentration of receptor in the cytosol was of the order of 1 fmol/mg protein and corresponded to 72 fmol/mg tissue DNA (that is, 100 and 20 times less than that in corresponding prostatic cytosol, respectively). In the adult female rat 2 days after castration, the concentration of receptor in the cytosol was 0.34 fmol/mg protein. Treatment with testosterone pellets (20 mg for 15 days) increased androgen receptor concentration significantly. In spite of the relatively low concentration of androgen-binding sites, the typical binding specificity of the androgen receptor and the regulatory effects of androgens on their own receptor support the possibility that some effect(s) of androgens upon skeletal muscles may be initiated directly at the cellular level through this receptor, a concept which is also in agreement with recently demonstrated in vitro effects of androgens on cultured myoblasts.