Influence of carbon dioxide enrichment, ozone and nitrogen fertilization on cotton (Gossypium hirsutum L.) leaf and root composition

The objective of this study was to test whether elevated CO₂], O₃] and nitrogen (N) fertility altered leaf mass per area (LMPA), non‐structural carbohydrate (TNC), N, lignin (LTGA) and proanthocyanidin (PA) concentrations in cotton (Gossypium hirsutum L.) leaves and roots. Cotton was grown in 14 dm³ pots with either sufficient (0·8 g N dm ^{? 3}) or deficient (0·4 and 0·2 g N dm ^{? 3}) N fertilization, and treated in open‐top chambers with either ambient or elevated ( + 175 and + 350 μ mol mol ^{? 1}) CO₂] in combination with either charcoal‐filtered air (CF) or non‐filtered air plus 1·5 times ambient O₃]. At about 50 d after planting, LMPA, starch and PA concentrations in canopy leaves were as much as 51–72% higher in plants treated with elevated CO₂] compared with plants treated with ambient CO₂], whereas leaf N concentration was 29% lower in elevated CO₂]‐treated plants compared with controls. None of the treatments had a major effect on LTGA concentrations on a TNC‐free mass basis. LMPA and starch levels were up to 48% lower in plants treated with elevated O₃] and ambient CO₂] compared with CF controls, although the elevated O₃] effect was diminished when plants were treated concurrently with elevated CO₂]. On a total mass basis, leaf N and PA concentrations were higher in samples treated with elevated O₃] in ambient CO₂], but the difference was much reduced by elevated CO₂]. On a TNC‐free basis, however, elevated O₃] had little effect on tissue N and PA concentrations. Fertilization treatments resulted in higher PA and lower N concentrations in tissues from the deficient N fertility treatments. The experiment showed that suppression by elevated O₃] of LMPA and starch was largely prevented by elevated CO₂], and that interpretation of CO₂] and O₃] effects should include comparisons on a TNC‐free basis. Overall, the experiment indicated that allocation to starch and PA may be related to how environmental factors affect source–sink relationships in plants, although the effects of elevated O₃] on secondary metabolites differed in this respect.