Comparisons were made of the sexual behaviour of sham-operated male hamsters and castrated males receiving testosterone, dihydrotestosterone or androstenedione (1-5 mg/week), or oil alone. Tests of short duration (10 min) were conducted at Week 3 (when the animals were sexually naive), Week 6 and Week 9. Sham-operated males showed marked increases in many elements of behaviour between Weeks 3 and 9, while castrated males receiving no androgen replacement showed marked decreases. Males receiving each of the three androgens showed marked increases in behaviour, but androstenedione-treated males showed less facilitation of sexual behaviour than controls. Dihydrotestosterone was as effective as testosterone. The three androgens were comparable in maintaining seminal vesicle weight after castration and in preventing the customary rise in pituitary and body weights. These data suggest that, unlike the situation in the rat, aromatization of androgens is unnecessary for the display of sexual behaviour in the hamster.