For all panels, mock – open circle, DC3000hrpA – closed circles, DC3000 – open squares. a, DC3000 decreases photosynthetic CO2 assimilation in saturating light (Asat). b, A/Ci curve showing the relationship between photosynthesis (A) and intercellular CO2 concentration (Ci) at 10 hpi. c, Maximum quantum efficiency of PSII (Fv/Fm), d, maximum light adapted quantum efficiency of PSII (Fv′/Fm′) and e, PSII operating efficiency (Fq′/Fm′) are decreased by DC3000 challenge and f, photochemical quenching of PSII (qL) is increased. g, NPQ transiently increases 4 hpi after DC3000 inoculation. h, DC3000 bacterial growth (inoculum of ∼0.5 × 108cfu ml−1) is restricted until 8 hpi. Asterisks show significant differences in bacterial growth (t-test, P < 0.05) from time 0 (mean ± s.d.; n = 6). i, False colour image of Fv/Fm at 16 hpi showing a decrease (green/yellow) after DC3000 challenge. Asterisks show inoculated leaves. j, False colour image of NPQ at 7 hpi showing the increase (blue) after DC3000 challenge. Asterisks show inoculated leaves. Photosynthesis values (a) are means ± s.d. (n = 4), and show that DC3000 differs significantly from DC3000hrpA and mock treatments from 8 hpi (*P < 0.001; two-way ANOVA). Chlorophyll fluorescence parameters (c–g) are means ± s.d. (mock and hrpA, n = 3; DC3000, n = 4). ANOVA with the least significant difference post hoc test and Bonferroni multiple comparison correction shows that DC3000 differs significantly from DC3000hrpA and mock treatments for all time points at and beyond the asterisk (P < 0.05).