It is more difficult to obtain a reliable assessment of antibiotic resistance in populations of aquatic bacteria than in those populations which are well characterized (e.g. bacteria of medical and veterinary significance). Factors which influence the results include the bacterial taxa involved, their site of origin and the methods and media used to isolate and subculture the bacteria, and to perform the sensitivity tests. Examples of these effects are provided. The resistance profiles obtained with populations of aquatic pseudomonads depend on the species composition of the population. Resistance patterns in aquatic bacteria varied with the site from which they were isolated; a higher incidence of resistance was recorded along shorelines and in sheltered bays than in the open water. The inclusion of antibiotics in the media employed for primary isolation increased the number of individual and multiple resistances recorded. A similar effect was observed with increased inoculum size in the sensitivity disc method but this could be reversed by raising the incubation temperature. The medium used to conduct the test also affected the results and many aquatic bacteria failed to grow on media such as Iso‐Sensitest Agar. It is recommended that the sensitivity disc method is adopted for aquatic bacteria because it permits interpretation of a wider range of response. Comparison of the incidence of antibiotic resistance in different habitats will remain meaningless, however, until comprehensive methods for the identification of bacteria are developed and the techniques used for sensitivity testing are standardized.