There is an evident decline in human fertility with age. The fundamental physiological question is whether the ovary, the uterus, or both are affected by the changes induced within the body by senescence. Based on clinical studies performed in in-vitro fertilization (IVF) laboratories, there is no doubt that age affects the oocyte. However, the effect of age on the remaining cellular components of the preovulatory follicle have yet not been fully elucidated. In addition, whether the ability of the uterus to have normal implantation of human embryos is age-related is also a matter of controversy. In this review, we analyse the results of our studies that were conducted in order to answer these relevant questions. In the first series of experiments we tested the hypothesis that the function of granulosa cells is affected in older patients (> or = 40 years). In these experiments, we cultured in vitro granulosa-luteal cells from 15 women aged < 40 years and 18 patients aged > or = 40 years undergoing IVF. Immunoreactive alpha-inhibin and progesterone were measured after 48 and 96 h in culture. The measurements showed a significantly higher secretion of inhibin (P < 0.01) after 48 h, and progesterone (P < 0.05) after 48 and 96 h, in the younger as compared to the older group. In addition, we tested the ability of the uterus to sustain normal implantation by an analysis of our oocyte donation programme. In the first approach, we divided all the transfer cycles according to the age of the recipient. We found no differences among groups with regard to pregnancy, implantation or abortion rates. We explored this fundamental question further by performing oocyte donation cycles in which oocytes from the same donor were distributed into a recipient aged < 40 years and another recipient of > or = 40 years. An analysis of 45 cycles in each group of recipients showed a significant (P < 0.05) increase in abortion rates in older as compared to younger patients. When the serum oestradiol concentrations were followed in those women who carried normal single ongoing pregnancies, we found that a significant increase in blood oestradiol concentrations (which probably represents the luteo-placental shift) was detected 2 weeks earlier in younger than in older patients, suggesting a defective vasculature of the uterus that could ultimately be the cause of an increased abortion rate in patients aged > or = 40 years. In conclusion, we firmly believe that senescence affects both the ovary (oocyte and granulosa cells) and the uterus, and these observations should be taken into account in order to advise our patients.