In 1982, 61 selections were taken from J.E. Quisenberry's T-25 drought tolerant nursery in Lubbock, Texas. These selections were then taken into a greenhouse and crossed reciprocally with three commercial cultivars, Early Cot 48, Lambright X-15-4, and Pima S-5. In 1985 the T-25 drought tolerant parents, the three commercial parents, and the resulting progeny from the reciprocal crosses were planted in a rainout shelter and canopy temperatures measured twice daily by a handheld infrared thermometer from 35 through 65 days after planting. A total of 427 entries were planted, because of the large number of entries replication was not possible.

The same process was followed in 1986 except that the number of drought tolerant parents were reduced to seven selections that had proven to be the extremes on the canopy temperature scale. Measurements were taken on four replications. Additionally, in 1986 an irrigated location was included in the study to provide a check on the reliability of the readings taken in the rainout shelter.

Preliminary results in 1985 showed significant canopy temperature differences between parents and crosses and between different Fl progeny indicating differences in water use. However, no differences were found between reciprocal crosses leading to the conclusion that cytoplasmic effects are not a critical factor when trying to breed for drought tolerance. Additionally, it was suspected that the variation among plants of a single line might be at least as great as the differences between lines.

Data collected on the reduced number of entries in 1986 corroborated earlier results and added additional evidence to support the theory that variability in canopy temperature is widespread even between plants that should be genotypical similar both in drought stressed and well watered locations.