The manner in which exposure is defined can affect the findings of reproductive studies of toxic exposures. The individual end points potentially examined, such as fetal loss, subfertility, and congenital malformations observed at birth, are on a continuum by severity of effect: the most extreme effects of the three being infertility because no pregnancy is possible, and the least extreme, congenital malformations recognized at birth. End points observed at birth are survivors of a long and complex process. The process yielding one of these adverse end points may result from a number of factors, including level of exposure could result in early fetal loss, whereas a lower one might result in a congenital malformation observed at birth. If the probability of a less severe end point falls due to increasing probability of more severe end points with increasing exposure, then a nontraditional dose-response relationship may be observed in the study of one type of outcome.