The study compares the response after gonadotrophin-releasing hormone agonist (GnRHa) and human menopausal gonadotrophin (HMG) stimulation for in-vitro fertilization (IVF) in patients with either one or two ovaries. The study group (group A) included 73 cycles in women who had unilateral oophorectomy before their IVF treatment and the control group (group B) included 988 cycles in women with two ovaries. Tubal disease was the sole cause for infertility in all cases. The two groups were similar in age and parity. The patients with one ovary required more ampoules of HMG (62.9 versus 48.9, P < 0.001), a longer induction period (13.5 versus 12.7, P < 0.01) and had significantly lower oestradiol concentrations on the day of human chorionic gonadotrophin (HCG) administration (5840 versus 6473 pmol/l, P = 0.035). They yielded fewer follicles (11.2 versus 13.1, P = 0.005), fewer oocytes (7.3 versus 9.1, P = 0.006) and produced fewer embryos (4.4 versus 5.1, P < 0.05). There was no difference in fertilization rate (60 compared with 58%), or pregnancy rate (25.8 compared with 27.1% per oocyte retrieval). Women with only one ovary responded less well to GnRH agonist/HMG stimulation than women who had both ovaries but pregnancy outcome was the same in both groups.