Abstract. According to recent competition theory, the population dynamics of phytoplankton species in monoculture can be used to make a priori predictions of the dynamics and outcome of competition for light. The species with lowest ''critical light intensity'' should be the superior light competitor. To test this theory, we ran monoculture experiments and competition experiments with two green algae (Chlorella vulgaris and Scenedesmus protuberans) and two cyanobacteria (Aphanizomenon flos-aquae and a Microcystis strain) in light-limited continuous cultures. We used the monoculture experiments to estimate the critical light intensities of the species. Scenedesmus had by far the highest critical light intensity. The critical light intensities of Chlorella, Aphanizomenon, and Microcystis were rather similar. According to observation, Aphanizomenon had a slightly lower critical light intensity than Chlorella and Microcystis. However, according to a model fit to the monoculture experiments, Chlorella had a slightly lower critical light intensity than Microcystis, which in turn had a slightly lower critical light intensity than Aphanizomenon. These subtle differences between observed and fitted critical light intensities could be attributed to differences in the light absorption spectra of the species. The competition experiments were all consistent with the competitive ordering of the species according to the fitted critical light intensities: Chlorella displaced all three other species, Microcystis displaced both Aphanizomenon and Scenedesmus, and Aphanizomenon only displaced Scenedesmus. Not only the final outcomes, but also the time courses of competition predicted by the theory, were in excellent agreement with the experimental results for nearly all species combinations.