A literature review of pesticide toxicity to aquatic plants was conducted in order to arrive at a set of species for use in preregistration testing. Criteria used in the selection process were: (1) existing database; (2) interspecies variability in response to pesticides; (3) availability of well-characterized plant cultures; and (4) ease of culture. Freshwater and marine algae and freshwater and estuarine macrophytes were included in the study. Results showed that the three practical criteria [1,3,4] all resulted in the selection of unicellular green algae (Chlorophyceae) for the freshwater algae tests. However, variability among freshwater species and classes was so great and so unpredictable that it became obvious that a species battery approach was required. The battery should consist of representatives of each of the major algal classes. Research into culture methods and alternative endpoints is required before species other than unicellular green algae can be routinely incorporated into species battery tests. The studies in the marine algal database had much more consistent methodology; therefore, it was possible to include relative sensitivity and variability as well as practical considerations in the selection of marine tests species. The golden-brown algae (Chrysophyceae) were generally the most sensitive. However, because variability in response was still high, we recommend a species battery for marine algae as well. Four species of golden brown alga, two diatoms (Bacillariophyceae) and two green algae, were chosen based on the four criteria. Macrophytes should be part of preregistration testing because they can be more sensitive than algae and because of their ecological importance. The database showed a great variety of test methods and very few laboratory methods. No species emerged as the most consistently sensitive among either freshwater or estuarine macrophytes. Lemna gibba and Lemna minor are fast growing, easy to culture, and available commercially; they are logical test species for effects of pesticide drift and surface films. However, they are not always very sensitive. Rooted macrophytes belong in a comprehensive test program; however, test methods have to be developed. One criterion not included in this study was ecological importance. We recommend that ecological importance become part of any selection process in order to facilitate extrapolation to the field during hazard assessments.