Since phytoplankton production is usually estimated from static incubations (fixed depths or light levels), a mesocosm study was performed to evaluate the significance of mixing depth, mixing intensity and load of humus of natural phytoplankton assemblages. Vertically rotated (dynamic) incubations usually gave higher results than static incubations in humus-rich water. Mixing intensity was of significant importance in one of 2years tested, but strong interaction effects with humus complicated the explanation. Differences in primary production between dynamic incubations did not fully reflect the received PAR dose, and increased humus and increased mixing depth increased the photo-assimilation efficiency. Different single-depth incubations did not provide a shortcut method to measure water-column primary production with high accuracy. Results diverged from theoretical estimates based on recent combined photo-biological and physical environmental models. The large variability in responses to mixing is supposed to reflect species-specific adaptations and pre-history regarding quantity (photons) and quality (spectral distribution) of the optical environment in an assemblage of different species. The proportional abundance of each species with its specific characters will therefore strongly influence bulk primary production. Due to such variable responses, clear guidelines for a best practice in primary production measurements cannot be given, based on the present results.