Following treatment, seedlings were incubated with fluorescent dyes in liquid media of the same composition used in the stress treatment. For the visualization of sodium, seedlings were incubated with 5 μM CoroNa Green-AM (Invitrogen) for 2 h. Where confocal planes of the centre plane were required, seedlings were incubated for 8 h on a filter paper soaked with media supplemented with 2.5 μM CoroNa Green-AM, before washing and confocal microscopy at excitation and emission wavelengths of 488 nm and 516 nm, respectively, as described by Oh et al. (2009) (link). Negative control pictures of roots incubated without Na+ ions are presented in Supplementary Fig. S3 at JXB online. For analyses of intracellular pH, 10 μM carboxyl SNARF-AM (Invitrogen) was applied in the presence of 0.01% pluronic acid (Invitrogen) for 2 h. The ratio of average fluorescence intensities collected from the acidic (570–590 nm) and basic (630–650 nm) components with a single excitation wavelength at 488 nm were analysed by ImageJ software (NCBI). pH values were calculated by in situ calibration (see Supplementary Fig. S4 at JXB online) as described by the supplier (Invitrogen, MP01270). To visualize calcium, 20 μM Fluo4-AM was loaded in the presence of pluronic acid for 2 h. In separate experiments, Calcein-AM (Invitrogen), a non-ion specific fluorescence dye with similar molecular weight,was substituted for Fluo4-AM to ensure the same efficiency of dye loading between Col3 and sos1-1 root cells. Excitation and emission wavelengths of 488 nm and 516 nm were used for both Fluo4 and Calcein dyes. Where indicated, 1 μg ml−1 propidium iodide (Invitrogen) was added to counterstain the cell wall and dead cells. To visualize membrane trafficking, 5 μM FM4-64 (Invitrogen) was loaded for 5 min and the seedlings were incubated for a further 30 min after the removal of FM4-64.