In order to investigate the relative impacts of increases in day and night temperature on tree carbon relations, we measured night-time respiration and daytime photosynthesis of leaves in canopies of 4-m-tall cottonwood (Populus deltoides Bartr. ex Marsh) trees experiencing three daytime temperatures (25, 28 or 31 °C) and either (i) a constant nocturnal temperature of 20 °C or (ii) increasing nocturnal temperatures (15, 20 or 25 °C). In the first (day warming only) experiment, rates of night-time leaf dark respiration (Rdark) remained constant and leaves displayed a modest increase (11%) in light-saturated photosynthetic capacity (Amax) during the day (1000–1300 h) over the 6 °C range. In the second (dual night and day warming) experiment, Rdark increased by 77% when nocturnal temperatures were increased from 15 °C (0·36 µmol m−2 s−1) to 25 °C (0·64 µmol m−2 s−1). Amax responded positively to the additional nocturnal warming, and increased by 38 and 64% in the 20/28 and 25/31 °C treatments, respectively, compared with the 15/25 °C treatment. These increases in photosynthetic capacity were associated with strong increases in the maximum carboxylation rate of rubisco (Vcmax) and ribulose-1,5-bisphosphate (RuBP) regeneration capacity mediated by maximum electron transport rate (Jmax). Leaf soluble sugar and starch concentration, measured at sunrise, declined significantly as nocturnal temperature increased. The nocturnal temperature manipulation resulted in a significant inverse relationship between Amax and pre-dawn leaf carbohydrate status. Independent measurements of the temperature response of photosynthesis indicated that the optimum temperature (Topt) acclimated fully to the 6 °C range of temperature imposed in the daytime warming. Our findings are consistent with the hypothesis that elevated night-time temperature increases photosynthetic capacity during the following light period through a respiratory-driven reduction in leaf carbohydrate concentration. These responses indicate that predicted increases in night-time minimum temperatures may have a significant influence on net plant carbon uptake.