During the grain filling period we followed diurnal courses in leaf water potential (~), leaf osmotic potential (~0), transpiration (E), leaf conductance to water vapour transfer (g) and microclimatic parameters in field-grown spring barley (Hordeum distichum L. cv. Gunnar). The barley crop was grown on a coarse textured sandy soil at low (50 kg ha -~) or high (200 kg ha -1) levels of potassium applied as KCI. The investigation was undertaken at full irrigation or under drought. Drought was imposed at the beginning of the grain filling period.Leaf conductance and rate of transpiration were higher in the flag leaf than in the leaves of lower insertion. The rate of transpiration of the awns on a dry weight basis was of similar magnitude to that of the flag leaves. On clear days the rate of transpiration of fully watered barley plants was at a high level during most part of the day. The transpiration only decreased at low light intensities. The rate of transpiration was high despite leaf water potentials falling to rather low values due to high evaporative demands. In water stressed plants transpiration decreased and midday depression of transpiration occurred. Normally, daily accumulated transpirational water loss was lower in high K leaves than in low K leaves and generally the bulk water relations of the leaves were more favourable in high K plants than in low K plants.The factorial dependency of the flag leaf conductances on leaf water potential, light intensity, leaf temperature, and leaf-to-air water vapour concentration difference (AW) was analysed from a set of field data. From these data, similar sets of microclimatic conditions were classified, and dependencies of leaf conductance on the various environmental parameters were ascertained. The resulting mathematical functions were combined in an empirical simulation model. The results of the model were tested against other sets of measured data. Deviations between measured and predicted leaf conductance occurred at low light intensities. In the flag leaf, water potentials below -1.6 MPa reduced the stomatal apertures and determined the upper limit of leaf conductance. In leaves of lower insertion level conductances were reduced already at higher leaf water potentials. Leaf conductance was increased hyperbolically as photosynthetic active radiation (PAR) increased from darkness to full light. Leaf conductance as a function of leaf temperature followed an optimum curve which in the model was replaced by two linear regression lines intersecting at the optimum temperature of 23.4°C. Increasing 206 L6sch et al. leaf-to-air water vapour concentration difference caused a linear decrease in leaf conductance. Leaf conductances became slightly more reduced by lowered water potentials in the low K plants. Stomatal closure in response to a temperature change away from the optimum was more sensitive in high K plants, and also the decrease in leaf conductance under the influence of lowered ambient humidity proceeded with a higher sensitivity in high K plants. Thus, und...