Six cotton cultivars were evaluated in irrigated and non-irrigated field plots for differences in root and shoot traits, at seedling and mature stages, leaf canopy temperature, boll set, lent yield, and earliness. Four of the six cultivars were from the multi-adversity resistant (MAR) cotton improvement program. The two non-MAR cultivars were Deltapine 41 and paymaster 303. Root and shoot traits were measured at the seedling stage prior traits were measured at the seedling stage prior to irrigations and at plant maturity. Leaf canopy temperatures were measured weekly over a 4-week period, during peak flowering, with a hand-held infrared thermometer.

the irrigated treatment received 12 inches of water (six furrow irrigations of water (six furrow irrigations of approximately 2 acre inches each) in addition to 1.72 inches of rainfall from June 19 to August 30, 1987. The non-irrigated treatment received 1.72 inches of rainfall during the 73 day period.

Significant differences were observed among cultivars at the seedling stage for taproot length, number of lateral roots, total root dry weight and shoot dry weight. Differences also existed among cultivars for canopy temperature, root traits at maturity, total number of mature bolls, lint yield, and earliness. The MAR cultivars, TX-CABU'CS and Tamcot CD3H produced the greatest number of mature bolls and highest lint yield under irrigation. TX-CABU'CS also had the highest leaf canopy temperature under both irrigation and non-irrigation.

A stepwise multiple regression variable selection analysis indicated that seedling root-to-shoot ratio and taproot length explained 97% of the variability (R²) in total lint yield for the non-irrigated treatment. Results suggest that the early establishment of a healthy, vigorously growing seedling is important for increased yields when the plant is grown under non-irrigated, water stressed conditions.

In the irrigated treatment, a high leaf canopy temperature on july 31 explained 89% of the variability in total lint yield. The greatest amount of variability in yield was explained by a high leaf canopy temperature on July 31, a high root-to-shoot ratio at maturity, and a low seedling total root weight. The positive association of root-to-shoot at maturity with yield implies that cultivars which posses more extensive root systems may achieve higher yields by more surface area for water uptake. The relationship of a low seedling total root dry weight with lint yield was similar to the results of the non-irrigated treatment; where the establishment of a vigorous, healthy seedling is associated with higher yields. Once the healthy plant is established, the higher yielding cultivars appear to have a higher water use efficiency, due to an increased root distribution and decreased water loss through transpiration.

Plant traits under irrigation were then used to explain variability in non-irrigated lint yield. The greatest amount of variability (R² = 0.96) was explained by a long taproot length, a high leaf canopy temperature on July 31, and a high proportion of lateral roots in the total root dry weight. This should provide more useful information for identifying heritable traits associated with drought tolerance, since the irrigated treatment would be less influenced by environment due to the controlled conditions. The results of this study confirm the relationship of higher yields with linger, more extensive root systems for increasing the water absorption suggests that some cultivars have the inherent ability to minimize water loss through transpiration.