Drought and salinity are two widespread environmental conditions leading to low water availability for plants. Low water availability is considered the main environmental factor limiting photosynthesis and, consequently, plant growth and yield worldwide. There has been a long‐standing controversy as to whether drought and salt stresses mainly limit photosynthesis through diffusive resistances or by metabolic impairment. Reviewing in vitro and in vivo measurements, it is concluded that salt and drought stress predominantly affect diffusion of CO₂ in the leaves through a decrease of stomatal and mesophyll conductances, but not the biochemical capacity to assimilate CO₂, at mild to rather severe stress levels. The general failure of metabolism observed at more severe stress suggests the occurrence of secondary oxidative stresses, particularly under high‐light conditions. Estimates of photosynthetic limitations based on the photosynthetic response to intercellular CO₂ may lead to artefactual conclusions, even if patchy stomatal closure and the relative increase of cuticular conductance are taken into account, as decreasing mesophyll conductance can cause the CO₂ concentration in chloroplasts of stressed leaves to be considerably lower than the intercellular CO₂ concentration. Measurements based on the photosynthetic response to chloroplast CO₂ often confirm that the photosynthetic capacity is preserved but photosynthesis is limited by diffusive resistances in drought and salt‐stressed leaves.

The conductance for CO(2) diffusion in the mesophyll of leaves can limit photosynthesis. We have studied two methods for determining the mesophyll conductance to CO(2) diffusion in leaves. We generated an ideal set of photosynthesis rates over a range of partial pressures of CO(2) in the stroma and studied the effect of altering the mesophyll diffusion conductance on the measured response of photosynthesis to intercellular CO(2) partial pressure. We used the ideal data set to test the sensitivity of the two methods to small errors in the parameters used to determine mesophyll conductance. The two methods were also used to determine mesophyll conductance of several leaves using measured rather than ideal data sets. It is concluded that both methods can be used to determine mesophyll conductance and each method has particular strengths. We believe both methods will prove useful in the future.