Field studies were conducted at Bushland, Texas, USA, in 2004 to examine usefulness of canopy temperature depression(CTD), the difference of air-canopy temperature, in screening wheat (Triticum aestivum L.) genotypes for yield underdryland and irrigated. Forty winter wheat genotypes were grown under irrigation and dryland. CTDs were recorded afterheading between 1 330 and 1 530 h on 6 clear days for dryland and 9 days for irrigation. Drought susceptible index (DSI)for each genotype was calculated using mean yield under dryland and irrigated conditions. Genotypes exhibited greatdifferences in CTD under each environment. The dryland CTDs averaged 1.33°C ranging from -0.67 to 2.57°C, and theaverage irrigation CTD were 4.59°C ranging from 3.21 to 5.62°C. A low yield reduction was observed under drylandconditions relative to irrigated conditions for high-CTD genotypes. CTD values were highly negatively correlated withDSI under dryland, and genotypes of CTDs = 1.3°C in dryland condition were identified as drought resistant. For 21genotypes classified as drought resistant by DSI, their CTDs were 1.68°C for dryland and 4.35°C for irrigation on average;for 19 genotypes classified as drought susceptible by DSI, average CTD was 0.94°C in dryland and 4.85°C in irrigation.The high-yield genotypes consistently had high CTD values, and the low-yield ones had low CTD values for allmeasurements in dryland. After heading, genotypes maintained consistent ranking for CTD. Regression results for CTDand yield suggested that the best time for taking CTD measurement was 3-4 weeks after heading in irrigation but any timebefore senescence in dryland. Crop water stress index (CWSI) calculated from CTD data was highly correlated with CWSIcalculated from yield, which suggesting traditional costly CWSI measurement may be improved by using portable infraredthermometers. Most importantly, grain yield was highly correlated with CTD under dryland (R2 = 0.79-0.86) and irrigation(R2 = 0.46-0.58) conditions. These results clearly indicated grain yield and water stress can be predicted by taking CTDvalues in field, which can be used by breeding programs as a potential selection criterion for grain yield and droughtresistance in wheat, but a second study year is needed to confirm further.
