1. As in other fields of toxicology, pharmacokinetic variables are of considerable significance for the assessment of reproductive or developmental toxicity. 2. When assessing the possibility of an adverse effect of a chemical on male or female fertility, in the majority of the cases no information on the concentration at the target is available. This may cause problems in the case of a 'negative' finding. 3. Prenatal toxicity takes place within a complex 'two-compartment' system (mother: conceptus); pharmacokinetic variables in both systems determine the concentration at the target (the conceptus). These variables may be expected to vary greatly during the different stages of development. While the contribution of maternal pharmacokinetics can easily be assessed, the pharmacokinetics within the embryo/foetus will remain largely unknown in man. 4. While it is quite feasible in animal experiments to study the transfer of a given chemical to the conceptus in vivo at various stages of pregnancy it is almost impossible to obtain such information for most developmental stages in man, except at the perinatal period (by sampling cord blood). For this reason during most risk assessments exposure of the embryo or fetus can only be roughly assumed from maternal plasma concentrations. 5. A significant capacity of foetal tissues (including liver) for metabolic transformation of many chemicals (e.g. via cytochrome-P-450-dependent monooxygenases) develops only perinatally in the usual rodent animal species. But in primates (including man) the activity of several such monooxygenases is substantial in liver already at early foetal stages. Much less information is available on the metabolic capacity for xenobiotics in extrahepatic tissues of primate embryos and foetuses. 6. Since pharmacokinetic variables (e.g. half-life) often differ considerably in laboratory rodents and in man a risk assessment for man on the basis of doses is often greatly misleading. It seems more justified to predict toxic effects on the basis of plasma levels. This aspect is of great significance and must be taken into consideration since the results of many routine studies are not suitable for risk assessments because pharmacokinetic variables were not considered when planning experimental studies. 7. Factors affecting the transfer of chemicals via mothers milk to the neonate are largely known today. But solid data on the pharmacokinetics of xenobiotics with respect to transfer to the milk and ensuing concentrations in the neonate are still lacking for the majority of relevant chemicals.(ABSTRACT TRUNCATED AT 400 WORDS)