Inter-species correlations could be a useful tool for predicting toxicity and for establishing sensitivity ratios among species. In this paper, quantitative inter-specific chemical activity relationships (QICAR) for aquatic organisms were developed to verify if such an approach could be utilised for estimating toxicological data when no other information is available. Inter-specific toxicity relationships on fish, Daphnia and algae were performed for pesticides considering a large data set (more than 600 compounds) and grouping the data either on a functional (herbicides, fungicides and insecticides) or chemical class base. Good correlations were found between several fish species and they were improved by excluding, from the data set, highly specific compounds such as organophosphorus insecticides. Relationship between fish (rainbow trout) and Daphnia was significant for the whole data set, but clearly improves if congeneric classes of pesticides are considered. The most significant results were found for azoles (fungicides) and for all data set of pesticides with the exclusion of organophosphorus and carbamate insecticides. As expected, toxicity on algae does not correlate either with fish or with Daphnia on the whole data set, but excluding the classes acting specifically toward one organism (insecticides and several classes of herbicides), good relationships were found. The analysis of the data permits the conclusion that the specificity in the mode action of pesticides is the key parameter for expecting or not inter-specific relationships. By the relative specificity of action of a group of compounds towards two species, the probability of obtaining a QICAR for this group can be derived. In general, compounds acting with the same level of specificity towards two different species, have a higher probability of showing inter-specific relationships and the lower the specificity of the mode of action of the compounds (e.g. narcotics or less inert chemicals), then the stronger are the relationships.