Goats in Group A were pretreated for 9 days with a synthetic progestagen, administered via intravaginal sponge, and 1000 i.u. PMSG s.c. on Day 12 of the oestrous cycle. Goats in Group B had the same PMSG treatment, but not the progestagen pretreatment. Group C goats received a s.c. twice daily injection of a porcine FSH preparation (8 mg on Day 12, 4 mg Day 13, 2 mg Day 14 and 1 mg Day 15). Oestrus was synchronized in all animals by 50 micrograms cloprostenol, 2 days after the start of gonadotrophin treatment. The vaginal progestagen sponges were removed from Group A at the same time. Mean ovulation rate was slightly higher in FSH-treated than in the PMSG-treated animals, whereas the incidence of large follicles that failed to ovulate was significantly elevated in PMSG-treated animals in Group B. More goats in Groups A and B than in Group C exhibited premature luteal failure. Progestagen pretreatment appeared to suppress both follicular and luteal activity, as indicated by numbers of large non-ovulating follicles and by the magnitude and duration of elevated plasma oestradiol levels following PMSG stimulation, and by decreased plasma progesterone levels before and after PMSG treatment. Oestrogenic response to FSH was considerably less than that to PMSG, as indicated both by a considerably shorter duration of elevation of circulating oestradiol levels during the peri-ovulatory period, and by lower maximal oestradiol levels. Differences in the ovarian responses to PMSG and FSH may be attributed primarily to differences in the biological half-life of each preparation.