A total of 82 normogonadotrophic clomiphene-resistant anovulatory patients were treated with esogenous gonadotrophins according to a step-down dose regimen during 234 cycles. In 43 (18%) cycles co-treatment with gonadotrophin-releasing hormone analogues was applied, The initial dose was between 1.5 and 2.5 ampoules (75 IU follicle-stimulating hormone each) per day (dependent on body weight), and decreasing steps of 0.5 ampoules/day were based on sonographic findings, The overall ovulation rate was 91% (213 cycles), The median treatment period was 11 days and a total of 14 ampoules of gonadotrophin were needed, In 131 (62%) of the ovulatory cycles not more than one, and in 208 (98%) cycles not more than two, follicles greater than or equal to 16 mm were present on the day human chorionic gonadotrophin was given, A total of 37 pregnancies occurred of which two were twins and one was a triplet (multiple pregnancy rate 8%), The pregnancy rate per cycle was 17% and the cumulative pregnancy rate after 7 months was 47%. The abortion rate was 19%. There were four (1.7%) cases of mild ovarian hyperstimulation, of which none became pregnant. In conclusion, this study shows that the applied step-down regimen for gonadotrophin induction of ovulation can be a safe and effective treatment alternative for patients with clomiphene-resistant anovulation. The duration of ovarian stimulation and the amount of exogenous gonadotrophin required is limited, Pregnancy rates are comparable with those reported for step-up regimens, and a low incidence of complications (i.e. multiple gestation and ovarian hyperstimulation) was noted, Although data obtained from this non-comparative study appear favourable, a prospective comparative trial is mandatory to confirm and extend these observations.