Stomatal density tends to vary inversely with changes in atmospheric CO(2) concentration (C(a)). This phenomenon is of significance due to: (i) the current anthropogenic rise in C(a) and its impact on vegetation, and (ii) the potential applicability for reconstructing palaeoatmospheric C(a) by using fossil plant remains. It is generally assumed that the inverse change of stomatal density with C(a) represents an adaptation of epidermal gas conductance to varying C(a). Reconstruction of fossil C(a) by using stomatal density is usually based on empirical curves which are obtained by greenhouse experiments or the study of herbarium material. In this contribution, a model describing the stomatal density response to changes in C(a) is introduced. It is based on the diffusion of water vapour and CO(2), photosynthesis and an optimisation principle concerning gas exchange and water availability. The model considers both aspects of stomatal conductance: degree of stomatal aperture and stomatal density. It is shown that stomatal aperture and stomatal density response can be separated with stomatal aperture representing a short-term response and stomatal density a long-term response. The model also demonstrates how the stomatal density response to C(a) is modulated by environmental factors. This in turn implies that reliable reconstructions of ancient C(a) require additional information concerning temperature and humidity of the considered sites. Finally, a sensitivity analysis was carried out for the relationship between stomatal density and C(a) in order to identify critical parameters (= small parameter changes lead to significant changes of the results). Stomatal pore geometry (pore size and depth) represents a critical parameter. In palaeoclimatic studies, pore geometry should therefore also be considered.