Fossil evidence shows that stomata have occurred in sporophytes and (briefly) gametophytes of embryophytes during the last 400 m yr. Cladistic analyses with hornworts basal are consistent with a unique origin of stomata, although cladograms with hornworts as the deepest branching embryophytes require loss of stomata early in the evolution of liverworts. Functional considerations suggest that stomata evolved from pores in the epidermis of plant organs which were at least three cell layers thick and had intercellular gas spaces and a cuticle; an endohydric conducting system would not have been necessary for low-growing rhizophytes, especially in early Palaeozoic CO2-rich atmospheres. The ‘prestomatal state’ (pores) would have permitted higher photosynthetic rates per unit ground area. Functional stomata, and endohydry, permit the evolution of homoiohydry and the loss of vegetative desiccation tolerance and plants > 1 m tall. Stomatal functioning would then have involved maintenance of hydration, and restricting the occurrence of xylem embolism, under relatively desiccating conditions at the expense of limiting carbon acquisition. The time scale of environmental fluctuations over which stomatal responses can maximize carbon gain per unit water loss varies among taxa and life forms.