The 32P-postlabelling analysis provides a sensitive means for detecting pollution-related DNA adducts in aquatic organisms exposed to environmental carcinogens. However, the following factors need to be taken into consideration during the data interpretation: (1) species-specific, naturally occurring DNA modifications (or I-compounds) are found in aquatic organisms at levels which are highly season-dependent; and (2) many aquatic organisms, particularly lower invertebrates, cannot form DNA adducts from common pollutants such as polycyclic aromatic hydrocarbons (PAHs). The level of natural adducts is especially high in lower invertebrates, such as sponges and sea-urchins during their reproductive phase in the spring time (March/April): in subsequent months adducts were either undetectable or present at only trace levels. These invertebrates do not metabolize PAHs such as benzo[a]pyrene but readily biotransform aromatic amines such as 2-acetylaminofluorene to DNA-reactive forms. Pollution-related DNA adducts have been found in fish living in highly polluted rivers and marine sites and in carp exposed to an artificial Diesel-2/crude oil slick. In certain fish (English sole, brown bullheads, etc.) living in polluted environments, the formation of pollution-related DNA adducts has been correlated with an increased incidence of tumours. It is concluded that, while DNA adducts detected in aquatic organisms can be used for biomonitoring and detecting pollutants, there are several confounding factors that should be taken into consideration before one attempts to determine the type and concentration of carcinogenic pollutants present in aquatic environments.