The behavior of water stressed cotton (Gossypium hirsutum L.) is well documented, but our knowledge of traits which can be genetically manipulated to improve drought tolerance is incomplete. This study was conducted to determine which morphological and physiological factors lessen the effects of water stress on the yield of two short-season cultivars [i.e., TAMCOT HQg5 CHQg5) and G&P 74 ÷ (GP74)] with a common parent in ancestry. Plants were grown in a rain shelterlysimeter facility containing a Pedernales fine sandy loam soil (fine, mixed, thermic Udic Paleustalf) in 1990 and 1991 at Temple, TX. Water stress was imposed by replenishing a fraction of the water lost to evapotranspiration, beginning about 78 d after emergence. HQ95 and GP74 did not differ in leaf area index (LA]) or in the rate of leaf area development before water stress was imposed. The rate of LAI decline and average LAI were similar between cultivars when water stress was imposed. HQg5 used more water and used it at a faster rate than GP74 when water stressed. HQg5 produced more bolls and had higher yield under both well-watered and water stressed conditions than GP74 in each of the 2 yr. The largest difference in boll load between cultivars occurred on sympodia branches in the lower canopy, where HQ95 had 37, 60, and 182% more bolls than GP74 when plants received 0, 50 or 75, and 100% of the depleted soil water. Whether well watered or water stressed, individual boll weight did not differ between the two cultivars. However, the harvest index and the production efficiency of bolls (i.e., bossl pr unit leaf area) of HQ95 was consistently higher than GP74 for all water regimes. On average, HQ95 allocated 21% more dry matter to yield and produced 32% more bolls per square meter than GP74. While differences in yield between the cultivars mirror harvest index, large differences in boll production efficiency suggested that the intrinsic photosynthetic capacity of HQ95 may be greater than GP74. C OTa~ON is grown under a wide range of climatesfrom the humid sub-tropics of South Carolina to the semi-arid desert of California. Yet, water supply remains a critical limitation of yield. Genetic variation in yield and water-use efficiency has been reported for cotton subjected to water deficits (Quisenberry and McMichael, 1991; Cook and EI-Zik, 1993). Also, the effects of water deficits and relationships between cotton growth and water requirements are well documented (Grimes and EI-Zik, 1982; Morrow and Krieg, 1990). However, a fundamental understanding of the intrinsic factors which could improve the ability of cotton to withstand water deficits remains to be developed. We know that cotton fruit production and retention

Volume, 36, no. 4, p. 925: The units of y-axis of Fig. 3 were incorrectly printed as g[p.mol C0d(m2s)]; they should have been g [mol COd(m2s)]. The corrected figure is printed below.

No abstract

Water stress reduces net CO2 assimilation (A) and yield of cotton (Gossypium hirsututn L.), but our knowledge of the physiolog3, of water stress on A and assimilation capacity is incomplete. Experiments were conducted in a rain shelter-lysimeter facility in 1990 and 1991 to determine if the yields of two short-season cotton cultivars with common ancestry, TAMCOT HQ95 (HQ95) and G&P'/4 + (GP74), resuited from intrinsic differences in A and assimilation capacity. Water stress was imposed by withholding 0, 50 or 75, and 100% of the depleted soil water after flowering. Results indicated that both stomatal and nonstomatal factors were important in controlling A. HQ95 bad higher A and g than GP74 over leaf water potentials (~/L) ranging from-1.0 to-3.2 MPa. Nonstomatai limitations to A were more important than stomatal factors when WL was >-1.5 MPa. Stomatal factors limited A when ~L was <-1.5 MPa for both cultivars. The initial slope (Si) and the maximum A at high ci (Am~) declined with increasing water stress for both cnitivars. The S~ was greater for HQ95 than GP74 over the range in ~L and suggest that HQ95 had higher ribulose-l,5-bisphosphate carboxylase-oxygenase activity than GP74. Increasing water stress reduced Amx equally in both cultivars. This suggests that electron transport processes for ribulose-l,5bisphosphate regeneration of the cultivars did not differ. Therefore, stomatal and nonstomatai CO2 assimilation processes are important in limiting A of water stressed cotton. Intrinsic differences in these processes enable some cotton cnitivars to better tolerate water stress. D ROUGHT is an important limitation to cotton production. Reduction in photosynthetic activity and increases in leaf senescence are symptomatic of water stress and adversely affect cotton yield (Krieg, 1981; Constable and Rawson, 1980; Marani et al., 1985). Other effects of water stress include reduced cell growth and enlargement, leaf expansion, assimilate translocation, and transpiration (Hsiao, 1973). Although the consequences of water stress are well known, our ability to manipulate genetically these processes and improve drought tolerance is limited. However, genetic differences exist in the rate of A for cotton. Karami et al. (1980) found that an okra leaf cotton genotype had significantly higher A under water stress than its isoline with "normal" leaf morphology. They attributed their results to differences in source-sink relationships, nonstomatal limitations, and leaf senescence between the genotypes. Pettigrew et al. (1993) attributed the higher A of cultivars with "okra" shaped leaves to a higher specific leaf weight (g/m 2 leaf area) and higher leaf chlorophyll concentrations compared to "normal" leaf cultivars. They hypothesized that the genotypic differences in A were due to a higher concentration of the photosynthetic apparatus per leaf, caused by in

Empirically based relationships between canopy minus air temperature (Tc–Ta) regressed on vapor pressure deficit (VPD) have been described as measures of crop water stress indices (CWSI) and indicators for irrigation scheduling. This study was conducted to determine seasonal and turfgrass species variation in empirical‐baseline functions. Empirical and energy‐balance CWSI functions also were compared to determine which was the most accurate estimate of CWSI over the range of turfgrass species and conditions studied. Field experiments were conducted to compare CWSI relationships derived during different climatic seasons and four different turfgrass species: buffalograss, Buchloë dactyloides (Nutt.) Engelm. cv. Texoka; common bermudagrass, Cynodon dactylon (L.) Pers. cv. Arizona common; St. Augustinegrass Stenotaphrum secundatum (Walter) Kuntze cv. Raleigh; and, tall fescue Festuca arundinacea Schreber cv. Falcon. Data were collected in midsummer of 1986 and late summer of 1987 from plots irrigated with a linear‐gradient irrigation system. The CWSI relationships were calculated from the two lowest canopy temperatures in each plot during 7 July to 1 Aug. 1986 and 30 Aug. to 11 Sept. 1987. Differences between CWSI baseline functions from 1986 and 1987 for common bermudagrass, buffalograss, and tall fescue were highly significant (P < 0.01). Mean values of net radiation, VPD, and wind speed also were significantly different (P < 0.01) for the seasons. Vapor pressure deficit usually accounted for more than 50% of the variability in Tc–Ta across seasons and turfgrass species. Using the energy‐balance method to calculate CWSI and comparing these values with empirical calculated CWSI values reduced the portion of index differences that were greater than 0.1.