The responses of phytoplankton to turbulent motions in the surface mixed layer can be measured to estimate the rate of vertical mixing. If the time scale for the response (photoadaptation) is shorter than that for vertical mixing, phytoplankton will exhibit a vertical gradient associated with adaptation to ambient light, whereas if mixing occurs with a time scale shorter than that of photoadaptation, the surface mixed layer will be uniform with respect to the photoadaptive parameter. To examine the physiological bases for a model of vertical mixing and photoadaptation, we grew the marine diatom Thalassiosira pseudonana (clone 3H) at three photon flux densities and subjected the cultures to reciprocal light shifts, measuring physiological and chemical changes over the following 10 h. Several parameters, easily measured in nature and attributable primarily to phytoplankton, responded to fluctuating light on different time scales. After cultures were exposed to relatively bright light, both the initial slope of the photosynthesis-irradiance curve and in vivo fluorescence were depressed on a time scale of less than an hour. Photosynthetic capacity was also reduced transiently, but recovered over many hours to a high level characteristic of an adapted state. First-order kinetics (the current model of choice for describing photoadaptation) reasonably described the rapid responses of phytoplankton to bright light, but other parameters (i.e. cellular chemical composition and photosynthetic capacity) changed as a result of unbalanced growth and required much longer to adapt from low to high light as compared to from high to low light. A logistic model of this adaptation is presented. The model suggests that hysteresis of adaptation during vertical mixing may have important consequences. The vertical distributions of photoadaptive properties in mixed layers not only reveal the rate of vertical mixing, but show how phytoplankton integrate environmental fluctuations.