Experiments were performed on attached leaves of potted Sambucus nigra plants of approximately 50 to 80 cm in size. Plants were drawn from cuttings and cultivated in 40-cm pots in a climatic chamber at a PPFD of 220 μmol m⁻² s⁻¹ (16-h light/8-h dark) and a temperature of 20°C. Plants were amply supplied with water and nutrients. Stomatal movements were observed on mature leaves in a gas-exchange chamber designed for simultaneous measurement of CO₂-water gas exchange and microscopic observation of stomatal movements under controlled light, humidity, temperature, and CO₂ conditions (Kaiser and Kappen, 2001). Temperature, leaf-to-air mole fraction of water vapor (Δ_W) and [CO₂] in the cuvette were set to 20°C to 22°C, 2 mmol mol⁻¹, and approximately 360 μmol mol⁻¹, respectively. Irradiance (500 μmol m⁻² s⁻¹, 16-h light/8-h dark) was provided by a fiber-optic illuminator (Kaltlicht-Fiberleuchte FL-400 with Spezial Fiberoptik 400-F; Walz). After mounting the leaf in the gas-exchange cuvette, the plant was allowed to adjust to measuring conditions for at least 24 h. Leaves were fixed with the adaxial side to a Perspex plate with double-sided transparent adhesive tape (Tesa 56661–2; Tesa) to allow micromanipulation. Subsequently, the plate was mounted inside the cuvette, which allows observation of the lower leaf surface with a long-distance microscope lens (50×) led through the bottom of the gas-exchange cuvette (Kaiser and Grams, 2006). The microscope (Axiovert 25CFL; Zeiss) is mounted on a motorized translation stage, which allows repositioning of samples of selected stomata. Digital images of stomata were recorded with a video camera, digitized, and stored for subsequent measurement of aperture with custom image analysis software. The aperture of oil-treated pores can no longer be measured due to refraction of the oil. In these experiments, the area of the guard cell pair between the anticlinal walls was measured. This measure is linearly related to aperture when pores are open (Kaiser and Kappen, 2001) and can be taken as a surrogate measure of stomatal opening. Pore area or guard cell pair area were converted to circularity (c = width × 100/length) to allow comparison between differently sized stomata. The humidity control by a bypass compensation system was used to perform quick changes in air humidity from Δ_W = 2 to approximately 18 mmol mol⁻¹ by switching the humidity of the incoming air to lower humidity. Within approximately 90 s, Δ_W arrived at its new steady state (Fig. 3). One hour before increasing Δ_W, [CO₂] was reduced to approximately 60 to 70 μmol mol⁻¹, which is approximately the CO₂ compensation point for C3 plants (von Caemmerer and Farquhar, 1982). This avoids intercellular [CO₂] gradients due to locally suppressed CO₂ diffusion into the mesophyll. Stomatal responses to a reduction in air humidity were observed before and after the treatment with oil or adhesive foil on subsequent days, always beginning at the same time 4 h after illumination was switched on. Stomatal apertures were observed from at least 30 min before to 1.5 h after reduction of air humidity. Images were taken every 3 to 4 min if apertures changed fast, otherwise at longer intervals of up to 10 min.