This study evaluated the effects of estrogens and androgens on LH pulse frequency and amplitude in male subjects. To assess the role of estrogens we compared the serum LH pulse frequency and amplitude between 3 groups: 1) 8 agonadal subjects receiving no steroid treatment; 2) 6 agonadal subjects continuously treated with 50 αg ethinylestradiol/day; and 3) 17 eugonadal men. Mean serum LH levels and LH pulse amplitude were significantly lower in the agonadal subjects receiving estrogens (14.8 ± 5.4 (sd) U/L and 4.1 ± 1.5 U/L, respectively) than in the group of agonadal subjects not receiving sex steroid treatment (35.7 ± 8.4 U/L and 7.3 ± 2.0 U/L, respectively). The mean LH pulse frequency was 7.1 ± 1.5/7 h in the group not receiving sex steroid treatment and 6.0 ± 1.4/7 h in the group receiving estrogens (P NS). The LH pulse frequency in the eugonadal men (3.8 ± 1.3/7 h) was significantly lower than the frequency in both groups of agonadal subjects. The LH pulse amplitude was of the same magnitude in the estrogen-treated agonadal subjects and in eugonadal men (4.1 ± 1.5 U/L and 3.5 ± 1.2 U/L, respectively). The role of androgens was studied in 15 eugonadal male subjects (who presented for female role reassignment) by determining the effects of a novel nonsteroidal androgen receptor blocker, Anandron, on 1) basal and LH-releasing hormone (LHRH)-stimulated serum LH/FSH levels; 2) LH pulse fre-quency and amplitude; 3) sex steroid and sex hormone-binding globulin levels; and 4) serum PRL levels during an 8-week period. Basal and LHRH-stimulated LH levels and testosterone rose progressively during the first 6 weeks and reached a plateau thereafter, while estradiol levels continued to increase somewhat. The LH pulse amplitude and frequency had increased after 6 weeks (3.1 ± 0.6 vs. 4.5 ± 1.2 U/L and 4.4 ± 2.4 vs. 6.6 ± 1.1 pulses/7 h, respectively). Basal FSH levels were not affected while LHRH-stimulated FSH ievels progressively decreased from 2 to 6 weeks, after which they did not change. Along with the rise of estradiol levels an increase of sex hormone-binding globulin and PRL levels occurred. From these data we conclude that: 1) interruption of the negative feedback action of androgens on the hypothalamic-pituitary system is a slow process extending over 4-6 weeks; 2) antiandrogens increase the frequency of the hypothalamic LHRH pulse generator; 3) antiandrogens increase basal and LHRH-stimulated pituitary LH secretion; 4) androgens have little effect on the regulation of FSH secretion; the antiandrogen effect on the LHRH-stimulated FSH release might be due to an increase of estradiol levels or the increase of LHRH pulse frequency; and 5) estrogens reduce LH pulse amplitude.