The purpose of this research was to determine the magnitude of photorespiration in field-grown cotton (Gossypium hirsutum L.) as a function of environmental and plant-related factors. Photorespiration rates were estimated as the difference between measured gross and net photosynthetic rates.A linear increase in photorespiration was observed as air temperature increased from 22 to 40 degrees C at saturating photon flux density. At 22 degrees C, photorespiration was less than 15 per cent of net photosynthesis and very comparable to the dark respiration rate. At 40 degrees C, photorespiration represented about 50 per cent of net photosynthesis. Gross photosynthesis had a temperature optimum of 32 to 34 degrees C. Water stress, as indicated by Psi(L), did not alter the ratio of gross photosynthesis to net photosynthesis when the confounding effects of leaf temperature differences were accounted for in the data analyses. A reduction in both gross and net photosynthesis was apparent as Psi(L) declined from -2.0 megapascals indicating direct effects of water stress on the photosynthetic process. Photorespiration expressed as a proportion of net photosynthesis increased as water stress intensified.Cotton cultivars possessing a fruit load had significantly higher gross and net photosynthetic rates and lower photorespiration rates than did photoperiod-sensitive cotton strains without a fruit load. Within the fruiting types, which were genetically very similar, only minor differences were observed in the photorespiration:net photosynthesis ratios. However, in the photoperiod-sensitive strains, considerable genetic variability existed when photorespiration was expressed as a proportion of net photosynthesis. These results suggest that the kinetics of ribulose-1,5-bisphosphate carboxylase:oxygenase may be different and, thus, the possibility of genetically reducing photorespiration exists.