Demographers and physcians have, largely independently, been involved in analysis of fecundity and intrauterine pregnancy loss for several decades. Demographers were primarily responsible for precise measurement of the reproductive capacity of human beings, for definition of the notion of fecundability, and for significant progress in measurement of the risk of fetal mortality using the multiple entry life table. The apparent rate of fertility mortality was under 10%, but precise calculation using the life table yielded an estimate of 23.2 stillbirths and miscarriages per 100 pregnancies at 4 weeks of amenorrhea. Demographers recognized the contribution of breastfeeding to fertility control in traditional societies, and demographers determined the rate of sterility of cohorts of women at different ages. Increasing knowledge of genetics and the role of genetic anomalies in early spontaneous abortions and the advent of new techniques of "assisted procreation" in the late 1960s brought with them new questions of measurement and definition. Studies of the delay to conception in populations using no contraception indicate that effective fecundability, the rate of conceptions leading to live births, generally averages from 20-30% for the 1st cycle of exposure to risk and decreases regularly month by month thereafter. A form of equation for a probability distribution using just 2 parameters has been found useful for demonstrating the heterogeneity of couples; it shows that even after 12 or 24 months of unsuccessful attempts, the odds are in favor of eventual conception. Programs to assist conception imply hormonal stimulation of ovulation, attempts at fertilization in the optimal cycle phase, and other such steps. The success rate to which assisted fertility programs is compared should refer in this case to the probability of conception under optimal conditions, which may be closer to 50% than to 25%. The risk of intrauterine pregnancy loss increases rapidly with age and with previous pregnancy loss. The analogy with the unequal distribution of fecundability suggests use again of the same type of probability distribution. 1 of the few studies done of very early pregnancy loss suggested a rate of around 50%. Even assuming a lower rate, it is clear that only a minority of fertilizations actually lead to a live birth, and that fecundability and intrauterine mortality are in fact 2 aspects of a single phenomenon comprising effective fecundability. Evaluation of treatments for infertility depend on the odds of conception in the absence of treatment. If all couples treated are truly sterile, then each conception leading to a live birth represents a success of the treatment. But inclusion of hypofertile couples, a large proportion of whom would eventually conceive even without treatment, changes the evaluation criteria. In cases the effectiveness of various types of treatment should be evaluated in double-blind studies.