The modulation of pulsatile gonadotrophin release by endogenous ovarian steroids during the normal menstrual cycle may be involved in the initiation of the following menstrual cycle. The absence of this cyclical variation may, in some cases, be the cause of, or contribute to the cause of, amenorrhoea. To assess this the modulatory effect of gonadal steroid administration on the pulsatile release of gonadotrophins was studied in fourteen amenorrhoeic and four oligomenorrhoeic women. Pulsatility was assessed by samples collected at 10 min intervals during a 4 h morning period before and after treatment with either progesterone or micronized oestradiol or a sequential combination of both. Ten patients with intact positive oestrogen-gonadotrophin feedback responded to progesterone treatment by both a significant reduction in LH pulse frequency, from a mean of 4.1 to a mean of 2.1 pulses within the 4 h study period and an increase in pulse amplitude. Progesterone therapy did not affect mean LH concentrations but there was a significant reduction in mean FSH concentrations. In the eight patients with absence of positive feedback, none of the treatment regimes elicited significant changes in LH release. No definable FSH pulses were detected before or after treatment in either group. Both the changes in LH and FSH concentrations and their release observed in this study, support the concept that cycle initiation may be related to a reduced pituitary exposure to LHRH associated with elevated progesterone concentration in the luteal phase of the cycle. This selectively induces FSH synthesis and storage. Release of this stored FSH may occur as a result of failure of the corpus luteum and falling progesterone concentrations.