If a mother is stressed or anxious while pregnant her child is more likely to show a range of symptoms such as those of attention deficit hyperactivity disorder, conduct disorder, aggression or anxiety. While there remains some debate about what proportion of these effects are due to the prenatal or the postnatal environment, and the role of genetics, there is good evidence that prenatal stress exposure can increase the risk for later psychopathology. Why should this be? In our evolutionary history it is possible that some increase in these characteristics in some individuals was adaptive in a stressful environment, and that this type of fetal programming prepared the child or group for the environment in which they were going to find themselves. Anxiety may have been associated with increased vigilance, distractible attention with more perception of danger, impulsivity with more exploration, conduct disorder with a willingness to break rules, and aggression with the ability to fight intruders or predators. This adaptation for a future dangerous environment may explain why stress and anxiety, rather than depression, seem to have these programming effects; why there is a dose–response relationship with prenatal stress from moderate to severe and it is not only toxic stress that has consequences; why not all children are affected and why individual children are affected in different ways; and why the outcomes affected can depend on the sex of the offspring. An evolutionary perspective may give a different understanding of children in our society with these symptoms, and suggest new directions for research. For example, there is some evidence that the type of cognitive deficits observed after prenatal stress have specific characteristics; these may be those which were adaptive in a past environment.