The rates of transpiration from the upper and lower surfaces of leaves of Gossypium hirsutum, Xanthium stnrumarium, and Zes mays were compared with the rates at which helium diffused across those leaves. There was no evidence for effects of CO2 concentration or rate of evaporation on the resistance to water loss from the evaporating surface ("resistance of the mesophyll wall to transpiration") and no evidence for any significant wall resistance in turgid tissues. The possible existence of a wai resistance was also tested in leaves of CommeHins commaunis and Tulipa gesnenana whose epidermis could be easily peeled. Only when an epidermis was removed from a leaf, evaporation from the mesophyll tissue declined. We conclude that under conditions relevant to studies of stomatal behavior, the water vapor pressure at the sites of evaporation is equal to the saturation vapor pressure.The stomata are considered to be the main sites of resistance to transpiration within the leaf. The bulk of this water is thought to evaporate from the mesophyll tissue, but an appreciable amount may also come from the inner walls of epidermal cells bounding the substomatal cavities (e.g. 17). The proportioning between the two currents may depend on magnitude and distribution of small temperature gradients within the leaf (6, 23). The possibility that the mesophyll may also offer a resistance to evaporation of appreciable magnitude has been debated for more than 70 years (16,23). Various authors (e.g. 3) have concluded that the characteristics of the liquid flow system do not directly affect the rates of evaporation from leaves, but have ignored the effects of an internal cuticle which is reported to line the mesophyll cell walls (24). The hydraulic permeability of this cuticle is unknown but a value of 10-" msec-' bar-', which has been determined for storage tissue (1), is eluivalent to a resistance to diffusion of water vapor of 2 sec cmu-when the area of the mesophyll surface is 10 times that of the leaf (13). Several authors (10, 13, 15, from the outside air across the epidermis to the intercellular spaces (18) would be overestimated to the same degree. As a result of this, the intercellular CO2 concentration would be underestimated. Kaplan (14) found that the wall resistance to transpiration in A triplex halimus in air containing 300 ,u C02/1 air was much higher than at 0 ul 1-'. This raises the possibility that earlier measurements of wall resistance, which were made in air kept free of CO2 in order to reduce stomatal resistance, were underestimates. Jarvis and Slatyer (13) found that the wall resistance increased with increasing transpiration rate and with decreasing leaf water content.