In a prospective randomized study, the effect of two ovulation induction regimens on implantation rate of frozen/thawed pronucleate ova was investigated. Patients received either human menopausal gonadotrophin (HMG) or clomiphene/HMG. Ovulation induction was done on an individual basis using ultrasound and plasma 17 beta-oestradiol concentrations. Ovulation was induced with human chorionic gonadotrophin (HCG) when the leading follicle reached a diameter of 18 mm. Pronucleate ova had been frozen using the slow-freezing method of Lassalle et al. (1985) (Fertil. Steril., 44, 645-651) and were thawed in synchrony with the age of the endometrium. Both groups of patients were comparable for age, indication for in-vitro fertilization, pre-ovulatory 17 beta-oestradiol concentration, number of large follicles and number and quality of embryos transferred. The only difference found was that HCG was administered 1 day earlier in the HMG group compared to the clomiphene/HMG group (P < 0.01). Using univariate analysis, the pregnancy rate was higher in patients stimulated with HMG alone compared to those stimulated with clomiphene/HMG (27 versus 15% respectively; P < 0.03), when HCG was administered later in the menstrual cycle (P < 0.01) and when more and better quality embryos were transferred (P < 0.01). Using multivariate regression analysis, the influence of the stimulation on pregnancy rate was even more pronounced (P < 0.01) when the day of HCG administration and the number and quality embryos transferred were taken into account. Therefore, we conclude that HMG alone increases pregnancy rate compared to clomiphene/HMG during replacement cycles of frozen/thawed pronucleate ova. These data suggest that HMG results in a better endometrium receptivity for embryos.(ABSTRACT TRUNCATED AT 250 WORDS)