To examine whether luteal phase defect is, in part, causally related to insufficient gonadotrophin stimulation, we compared the relation of the increment of serum progesterone concentrations in response to human chorionic gonadotrophin (HCG) with its basal level at mid-luteal phase. Thirty-eight naturally cycling infertile women aged between 27-41 years old were evaluated for hormonal responses to HCG injection at the mid-luteal phase. We measured luteinizing hormone (LH), follicle stimulating hormone (FSH), oestradiol and progesterone concentrations, before and 1, 2 and 3 h after the administration of HCG (5000 IU, i.m.) 7 days after ovulation verified by ultrasonography. Eleven out of 38 women exhibited progesterone concentrations below 10 ng/ml (low progesterone group), and those remaining showed progesterone concentrations of > or = 10 ng/ml (normal progesterone group). The basal LH, FSH and oestradiol concentrations were essentially the same in both groups. Progesterone concentrations rose significantly 1 h after the injection and levelled off thereafter. The increment of progesterone concentrations at 1 h in the normal progesterone group was 5.7 ng/ml on the average, whereas that in low progesterone group was 1.1 ng/ml. Furthermore, the percentage increase in progesterone concentrations at 1 h in the normal progesterone group was significantly greater than that in the low progesterone group. Both groups equally exhibited significant but marginal increases in oestradiol concentrations 1 h after the injection. LH and FSH concentrations at 3 h decreased significantly in both groups. In summary, HCG readily stimulates progesterone production in normally functioning corpus luteum whereas its stimulatory effect is minimal on malfunctioning corpus luteum. This suggests that luteal phase defect is not caused by inadequate gonadotrophin stimulation and, therefore, does not benefit from HCG administration.