Competitive abilities for inorganic nutrients of five Amazonian phytoplankton species (Closterium libellula, Trachelomonas hispida, Navicula sp., Pinularia subcapitata and Lyngbya sp.) were tested under batch culture conditions. Data were collected from monocultures in order to fit models of growth rate (Monod-like for nitrogen, and Droop- like for internal phosphorus) and nutrient uptake (Monod-like in every case). A multinutrient interspecific competition model was built, and its accuracy was tested with data from a competition experiment using all the species together in batch culture. Competitive ability for nitrate differed quite a lot among these species. However, competitive abilities for phosphate storage did not show great variation according to Q_min values. Winners of competition resulted to be those species with lower K_s values for nitrate (P. subcapitata and Lyngbya sp.) and among these, the model showed that the best competitor for phosphorus would dominate (Lyngbya sp.). This shows that phosphorus storage was the most successful strategy among a group of the two best competitors for nitrate. However, in the competition experiment, both species were apparently coexisting at the end. Since model predicted dominance of Lyngbya sp. alone, it is most likely that the duration of the experiment was not enough to show exclusion of P. subcapitata. Alternative explanations would be that some remineralization of nutrients could have occurred, or that some unknown chemical interactions (allelopathy) were influencing our results.