A series of 4 experiments was performed to study the relative contribution of androgens and estrogens in the activation of sexual behavior in castrated male quail. The synthetic androgen methyltrienolone (R 1881) which is not metabolized in androgen target tissues activated sexual behavior in castrated birds and at the dose level of 0.5-1 mg/day/animal had the same potency as testosterone (T). However R 1881 was much more active than T in the induction of cloacal gland growth and activation of crowing, two typically androgen-dependent responses. This suggests that sexual behavior is not controlled by exactly the same mechanism as crowing or cloacal gland growth. In another experiment, estradiol (E2) alone activated sexual behavior but it is only at very high doses which had clear toxic effects that a significant behavioral activation could be observed. This questions the role of E2 as the physiological agent stimulating copulation in intact birds unless it is assumed that centrally administered E2 would be much more active compared to peripheral E2 which is exposed to a very intense peripheral catabolism. In the last two experiments, a clear synergism could be detected between 5 alpha-dihydrotestosterone (5 alpha-DHT) and E2 in the activation of sexual activity and doses of hormones could be defined which had almost no activity by themselves but significantly stimulated sexual behavior when given simultaneously. It was however impossible to define a hormonal treatment with T metabolites which restored behavior to its precastration level, a result very easily achieved with T treatments. Taken together, these data suggest that activation of sexual behavior in quail does not depend only on E2, nor 5 alpha-DHT nor even on their combined action. Considering that specific T receptors which probably do not bind 5 alpha-DHT are present in the brain, it would seem justified to reconsider the possible role played by T itself in the activation of behavior.