Androgen receptors in crude and partially purified 105,000 X g supernatant fractions from rat testis, epididymis, and prostate were studied in vitro using a charcoal adsorption assay and sucrose gradient centrifugation. Androgen metabolism was eliminated during receptor purification allowing determination of the kinetics of [3H]-androgen-receptor complex formation. In all three tissues, receptors were found to have essentially identical capabilities to bind androgen, with the affinity for [3H] dihydrotestosterone being somewhat higher than for [3H] testosterone. Equilibrium dissociation constants for [3H] dihydrotestosterone and [3H] testosterone (KD = 2 to 5 X 10(-10) M) were estimated from independently determined rates of association (ka congruent to 6 X 10(7) M-1 h-1 for [3H] dihydrotestosterone and 2 X 10(8) M-1 h-1 for [3H] testosterone) and dissociation (t 1/2 congruent to 40 hr for [3H] dihydrotestosterone and 15 h [3H] testosterone). Evaluation of the effect of temperature on androgen receptor binding of [3H]testosterone allowed estimation of several thermodynamic parameters, including activation energies of association and dissociation (delta H congruent to 14 kcal/mol), the apparent free energy (delta G congruent to -12 kcal/mol), enthalpy (delta H congruent to -2.5 kcal/mol), and entropy (delta S congruent to 35 cal col-1 K-1). Optimum receptor binding occurred at a pH of 8. Receptor stability was greatly enhanced when bound with androgen. Receptor specificity for testosterone and dihydrotestosterone was demonstrated by competitive binding assays. The potent synthetic androgen, 7 alpha, 17 alpha-dimethyl-19-nortestosterone, inhibited binding of [3H] testosterone or [3H] dihydrotesterone nearly as well as testosterone and dihydrotestosterone while larger amounts of 5 alpha-androstane-3alpha, 17 beta-diol and nonandrogenic steroids were required. Sedimentation coefficients of androgen receptors in all unfractionated supernatants were 4 and 5 to 8 S. Differences in sedimentation coefficients were observed following (NH4)2SO4 precipitation which did not influence the binding properties of the receptors. These results, together with measurements of3alpha/beta-hydroxysteroid oxidoreductase activity in vitro, suggest that organ differences in receptor binding of [3H] dihydrotestosterone and [3H] testosterone in vivo result from relative differences in intracellular concentrations of these androgens rather than from differences in receptor affinities.