Agronomy Journal Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 17 June 2005
Published in Agron J 97:1062-1071 (2005)
DOI: 10.2134/agronj2004.0152
© 2005 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moinuddin,
Right arrow Articles by Reynolds, M. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Moinuddin,
Right arrow Articles by Reynolds, M. P.
Agricola
Right arrow Articles by Moinuddin,
Right arrow Articles by Reynolds, M. P.
Related Collections
Right arrow Water Stress
Right arrow Crop Physiology & Metabolism

Drought Stress

Osmotic Adjustment in Wheat in Relation to Grain Yield under Water Deficit Environments

Moinuddina,*, R. A. Fischerb, K. D. Sayrec and M. P. Reynoldsc

a Plant Physiol. Lab., Potash Res. Inst. of India, Sector-19, Dundahera, Gurgaon-122016, Haryana, India
b Aust. Cent. for Int. Agric. Res., GPO Box 1571, Canberra, ACT 2601, Australia
c Int. Cent. for Improvement of Maize and Wheat (CIMMYT), Lisboa 27, Apdo Postal 6-641, 0660, Mexico, D.F., Mexico

* Corresponding author (moinuddin202{at}rediffmail.com)

Received for publication June 5, 2004.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Six greenhouse and three field experiments were conducted at the International Center for Improvement of Maize and Wheat (CIMMYT), Mexico, on bread wheat (Triticum aestivum L.) genotypes to ascertain the role of osmotic adjustment (OA) in sustaining grain yield and its stability under water deficit conditions. Under simulated water deficit conditions of the greenhouse and field, the genotypes differed considerably regarding OA, ranging from 0.31 to 0.86 MPa and from 0.60 to 0.99 MPa, respectively. When the mean values of OA across the six greenhouse experiments were regressed against those of grain yield obtained at different moisture levels of the line source, the correlation coefficient value increased with the increase in moisture stress, turning positively significant (P < 0.001) at the highest water deficit level. Similarly, osmotic potential at full turgor (OP100) and turgor potential (TP), measured in the greenhouse, were positively correlated (P < 0.05) with grain yield at the highest line source water deficit level. Osmotic adjustment and grain yield, both measured in the simulated water deficit condition in the field, were also positively correlated (P < 0.05). Besides, OA maintained yield stability through maintenance of turgor under water deficit during reproductive period of crop growth. The heritability of OA, OP100, and TP, computed by pooled analysis of variance across the six greenhouse experiments, was 0.74, 0.73, and 0.79, respectively. The results indicated that OA, as well as OP100 and TP, could be used as screening tools for drought-resistant bread wheat genotypes in the greenhouse. This study also demonstrated the appropriate greenhouse screening methodology in this regard.

Abbreviations: DSI, drought susceptibility index • LS1, first line source • LS2, second line source • OA, osmotic adjustment • OP, absolute osmotic potential • OP100, osmotic potential at full turgor • TP, turgor potential • WP, water potential • YSI, yield stability index


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
OSMOTIC ADJUSTMENT results in an active accumulation of solutes within the plants in response to a lowering of soil water potential (WP) under water deficit (Turner and Jones, 1980). In general, OA is achieved by absorbing ions (e.g., K+, Na+, Ca2+, Mg2+, Cl–, NO3–, SO4–, and HPO4–) or by accumulating organic solutes (e.g., free amino acids, sugar alcohols, quaternary ammonium compounds, and sugars). As a consequence, the osmotic potential of the cell is lowered, which in turn, attracts water into the cell and, thereby, tends to maintain its turgor (Morgan, 1984). Accumulation of solutes in roots leads to lowering of the osmotic potential of the root, which maintains the driving force for extracting soil water under water deficit condition (Wright et al., 1983b). Thus, OA helps plants to perform better in drought in terms of growth and productivity through maintaining turgor and water supply to the plant, which thereby maintains a comparatively higher photosynthetic rate and growth (Morgan and Condon, 1986; Ludlow and Muchow, 1990; Blum et al., 1999; Subbarao et al., 2000b). Occurrence of OA at sensitive crop reproductive stages has been reported to play a constructive role against floral abortion (Wright et al., 1983b), which results in maintaining grain number under water deficit (Inuyama et al., 1976; Wilson and Eastin, 1982; Leport et al., 1999; Moinuddin and Renu Khanna-Chopra, 2004). Additionally, OA has also been claimed to facilitate a better translocation of preanthesis carbohydrate reserves to the grain during the grain-filling period (Morgan, 1980; Piearce and Raschke, 1980; Subbarao et al., 2000b).

In fact, OA has been reported as an important drought adaptation mechanism in many crop plants (Ludlow and Muchow, 1990). A positive relationship between OA and grain yield under water deficit has been shown in grain sorghum [Sorghum bicolor (L.) Moench] (Tangpremsri et al., 1995), wheat (Triticum aestivum L.) (Morgan et al., 1986; Blum et al., 1999), pea (Pisum sativum L.) (Rodriguez-Maribona et al., 1992), and pigeonpea [Cajanus cajan (L.) millsp] (Subbarao et al., 2000a). A significant increase in seed yield of a group of genotypes with high OA over that with low OA has also been reported under water deficit condition in different crops (Ludlow et al., 1990; Morgan et al., 1991; Morgan, 1995; Moinuddin and Renu Khanna-Chopra, 2004).

Among wheat genotypes, significant differences regarding OA capacity exist and are associated with differences in crop growth and grain yield when water deficits in the soil and atmosphere are large enough to cause substantial reductions in plant WPs. Using semi-isolines for OA, Morgan and his associates (Morgan, 1983, 1995; Morgan and Condon, 1986; Morgan et al., 1986) proved that OA could be used as a selection criterion for screening wheat genotypes under water deficit. The present investigation was performed to ascertain (i) if there is a considerable range regarding phenotypic variability of OA among bread wheat genotypes, (ii) if the diverse bread wheat genotypes (instead of bread wheat isolines for OA) could be distinguished on the basis of OA in the greenhouse as well as field under simulated water deficit environments, and (iii) if OA could play a positive role in yield stability under water deficit. Besides, it was aimed at exploring if OA, measured under simulated stressful conditions in the greenhouse, could be employed for selecting diverse bread wheat genotypes for water deficit conditions in the field as claimed by Morgan et al. (1986) using semi-isolines of bread wheat for OA. If the answer is yes, what could be the appropriate methodology to screen the diverse bread wheat genotypes on the basis of OA under simulated stressful conditions in the greenhouse?


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Six greenhouse and three field experiments were conducted at the International Center for Improvement of Maize and Wheat (CIMMYT), Mexico, between 1988 and 1990. The bread wheat genotypes included in the greenhouse and field studies were selected on the basis of their varying grain yield recorded in a drought trial conducted at CIMMYT during the 1988–1989 winter season. They also showed similar dates of anthesis. Eight of the 25 bread wheat genotypes tested in the greenhouse and field experiments (Table 1) were common in all the greenhouse experiments, constituting the base of most conclusions in this study. Table 2 shows the data regarding maximum and minimum temperature of the greenhouse recorded during the soil-drying cycle as well as during the entire crop life cycle. A summary of the methodology adopted in each greenhouse experiment is given in Table 3.


View this table:
[in this window]
[in a new window]
  		Table 1. Genotypes/pedigrees included in greenhouse and field experiments. Numbers 1–7 are from CIMMYT, 8–19 from a Joint Project of CIMMYT and ICARDA (International Center for Agricultural Research in Dry Areas, Syria), and 20–25 from the collaborating countries of CIMMYT (countries of origin are given in parentheses). Letters indicate (A) genotypes included in all the greenhouse experiments (selected out of 75 genotypes from Field Exp. 1); (B) genotypes included in Greenhouse Exp. 4, 5, and 6; and (C) genotypes included in Field Exp. 2 and 3.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Mean maximum and minimum temperature of the greenhouse recorded during the soil drying cycle and the whole crop life cycle.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Summary of the methodology employed in the greenhouse experiments.

 
Greenhouse Experiments
Experiments 1–3 were conducted with eight bread wheat genotypes while Exp. 4–6 were designed with 15 bread wheat genotypes, including the eight bread wheat genotypes employed in Exp. 1–3. Thus, eight bread wheat genotypes were common in all the greenhouse experiments. In all the greenhouse experiments, six plants were grown per pot. The pots, measuring 0.15 m diam., were filled with a mixture of peat, sand, and normal field soil (2:3:5). A mixture of urea and triple superphosphate (approximately equivalent to 150 kg N ha–1 and 50 kg P2O5 ha–1, respectively) was applied to each pot. In Exp. 1–3, half of the fertilizer dose was given before sowing and half 40 d after sowing while full fertilizer dose was applied before sowing in Exp. 4–6. All the greenhouse experiments were laid out with three replications, randomizing the genotypes completely within two moisture regimes, namely water stress and no water stress. Water stress was imposed by withholding of water at various growth stages by irrigating the plants daily with 50% of the total water evapotranspired during 24 h, which was calculated by weighing the whole pots daily at 0800 h. Soil-drying cycle was given for 15 d in Exp. 1 and 3–6 and for 8 d in Exp. 2. In Exp. 1, water stress was imposed during anthesis to early dough stage (Zadoks Scale 65–83). In Exp. 2, plants were made to experience water stress from medium milk to hard dough stage (Zadoks Scale 75–87). Water was withheld from the pots during medium milk to hard caryopsis stage (Zadoks Scale 75–93) in Exp. 3 while soil-drying cycle was extended from ear emergence to 50% anthesis (Zadoks Scale 50–68) in Exp. 4. On the other hand, soil-drying cycle was imposed at early vegetative stages in Exp. 5 and 6, at Zadoks Scale 21–39 and 13–30, respectively. Four leaves were measured together for water relation parameters. At the end of the soil-drying cycle, two flag and two penultimate leaves were selected in Exp. 1; four flag leaves were selected in Exp. 2–4 while four youngest fully developed leaves were sampled in Exp. 5 and 6 for water relation measurements. Water relation measurements were made during 0830 to 1030 h in Exp. 1–3 and during 1200 to 1400 h in Exp. 4–6 (Table 3). Control plants were kept in well-watered conditions throughout the soil-drying cycle.

Field Experiments
Experiment 1
Field Exp. 1 was conducted between 1988 and 1989 near Ciudad Obregon, Sonora (northwest of Mexico) at the CIANO (Centro de Investigasiones Agricola de Noroeste) Experiment Station (26° N, 109° W; elevation 40 m above sea level). The soil of the experimental field was coarse sandy clay, mixed montomorillonitic Typic Caliciorthid (USDA Soil Taxonomy), low in organic matter (<10 mg g–1), slightly alkaline in pH (7.7), and with adequate K fertility. Before seeding, 150 kg N ha–1 and 20 kg P ha–1, as urea and triple superphosphate, respectively, were broadcast to the plots. The seed rate was 100 kg ha–1. The line source system was based on the design employed by Hanks et al. (1976). On each side of the line, there was a block of 75 plots 15 m long and 1.2 m wide, arranged perpendicular to the sprinkler line such that the gradient in water supply was achieved along the length of the plot. Each plot comprised six rows 0.20 m apart. To minimize the edge effects, no additional gap was left between plots beyond the normal 0.20-m row spacing. The 75 bread wheat genotypes were sown in random order in each of the two replicate blocks. After a uniform presowing irrigation, subsequent irrigations were applied by the line source system when the 0- to 0.60-m soil profile near the line source was depleted to about 50% of its maximal available water content. All irrigations were applied under calm conditions usually in the morning between 0530 to 0800 h to avoid wind drift. A total of 10 irrigations were applied during the crop season. The amounts of water applied in each irrigation were measured with rain gauges distributed uniformly throughout the trial area. The amounts of total water applied (including rainfall and available water in the profile at sowing) are given in Table 4 along with the corresponding mean grain yield of 8 (Exp. 1–6) and 15 (Exp. 4–6) selected bread wheat genotypes obtained at five moisture levels of the line source. Discounting 0.5 m at either end of each plot, the remaining 14 m was divided into five equal subplots (Positions 1 to 5, most to least stressed, respectively), each measuring 2.8- by 1.2-m (3.36 m2) area. Out of the six rows per subplot, the central four rows were hand-harvested from all five moisture regimes soon after physiological maturity, sun-dried, and threshed. Thus, harvest area was 2.24 m2 per subplot.


View this table:
[in this window]
[in a new window]
  		Table 4. Mean grain yield obtained and total water applied (including rainfall) at five moisture levels of the line source experiment (Field Exp. 1).

 
Experiment 2
This trial was also conducted at CIANO, Mexico, between 1989 and 1990, aimed at comparing the yield response of bread wheat genotypes to preanthesis water stress. In this experiment, 28 bread wheat genotypes were tested. However, only 18 of them with similar dates of anthesis were selected for OA measurement (Table 1), having varying yield response under water deficit. Measurement of OA in this field experiment was aimed at looking for the correlation between OA and yield, both recorded in the field. The design of the experiment was split plot with three replications, considering moisture regimes as main treatments and the genotypes as subtreatments. Three irrigation regimes were supplied as described below:
  1. One sowing irrigation plus one supplemental irrigation after anthesis; the amount of total water applied including rainfall was 337 mm (preanthesis stress).
  2. Only one sowing irrigation; total water applied including rainfall was 202 mm (severe stress).
  3. One sowing irrigation plus five supplemental irrigations; total water applied including rainfall was 587 mm (control).

Soil composition, row spacing, seed rate, fertilizer dose, and its mode of application were kept same as in Field Exp. 1. Other agricultural practices to accomplish the experiment were kept standard.

Experiment 3
It was a double line source experiment conducted along with Exp. 2 between 1989 and 1990 at CIANO, Mexico. The two experiments were simultaneously conducted to compare the response of 20 bread what genotypes to water stress in two drought-screening methodologies employed in Field Exp. 2 and 3, the 20 genotypes being common in both the experiments. However, measurement for OA was made on the same 18 genotypes studied in Field Exp. 2 (Table 1). Osmotic adjustment, measured in Field Exp. 2, and the grain yield, obtained at various moisture levels of this double line source experiment, were used to work out a correlation between OA and grain yield, both recorded in simulated drought condition in the field. In this experiment, the moisture regimes were imposed by two line sources, lying in parallel, 15 m apart. Thus, the line source experiment layout, described by Hanks et al. (1976), was repeated in sequence (Fig. 1) . Plots were extended to 15 m on both the sides of each line source, and the layout of the subplots for each line source system was the same as in Field Exp. 1. After sowing, all the plots were uniformly irrigated. Later, the first line source (LS1) was activated every 10 d for 1 h. The second line source (LS2) was activated only after anthesis. Thus, the plots on one side of LS1 experienced the standard line source moisture gradient throughout the crop season. Plots between the two line sources had varying degrees of preanthesis stress as imposed by LS1 but had no water stress after anthesis due to activation of both the line sources together after anthesis. Plots on the other side of LS2 observed a uniform severe preanthesis stress, followed by varying degrees of postanthesis moisture stress as determined by the standard line source moisture gradient imposed by LS2. Seed rate, fertilizer dose and its mode of application, plot size, row spacing, irrigation procedure, and other agricultural practices were kept the same as in Field Exp. 1.



View larger version (10K):
[in this window]
[in a new window]
  		Fig. 1. Layout of the double line source experiment (Field Exp. 3). LS1 and LS2 represent the two line sources, lying in parallel 15 m apart. The 15-m block extended on each side of the two line sources comprised 75 plots arranged perpendicular to the sprinkler line. Each plot was 1.2 m wide carrying six rows. Numbers 1–5 represent the line source moisture levels. The moisture level nearest to the line source (Level 1) was assumed as the irrigated plot (control) as it received the maximum water applied by the line source. LS1 was activated every 10 d for 1 h throughout the crop season while LS2 started functioning similarly after anthesis.

 
Measurement of Leaf Water Relations Parameters
Quantification of the crop water stress was made by measuring the leaf water relations parameters, namely WP, absolute osmotic potential (OP), and TP in all the greenhouse experiments. A summary of the greenhouse experiments conducted is given in Table 3, showing the number of the bread wheat genotypes included, duration of soil drying cycle imposed, crop development stage, and the type of leaf chosen for the study.

Water potential was measured in three replicates by the use of a pressure chamber (Soil Moisture Equipments Corp., Santa Barbara, CA, USA) according to Scholander et al. (1965), using four leaves at a time. The same leaves were wrapped in polythene strips and then frozen at –20°C in a deep freezer. After thawing at room temperature ({approx}15 min), cell sap was expressed using a hand press, and the OP of the cell sap was measured with the help of a vapor pressure osmometer (model 5500, Wescor, Inc., Logan, UT, USA). The osmometer was calibrated with known concentrations (mmol kg–1) of NaCl solutions. These values were converted to pressure unit according to the equation:

where R is the gas constant (0.008314) and T is the temperature measured in the Kelvin scale (298 K in our measurements). The OP was corrected (OP + 0.1OP) for the dilution of symplastic sap by apoplastic water, assuming 10% apoplastic water (Kramer, 1983). Turgor potential was estimated by the difference of water and osmotic potential. To measure OP100, in the greenhouse experiments, detached leaves were rehydrated in distilled water for 4 h, wrapped in polyethylene strips, and then frozen at –20°C. It was made sure that after rehydration, the WP of the leaf was at or near zero. In field condition, random sampling of four flag leaves was done from each treatment plot at anthesis. After rehydration for 4 h, the leaves were wrapped in polyethylene strips and frozen in dry ice kept in a thermocol box for the determination of OP100. Both in greenhouse and field condition, OA was determined as the difference of OP100 measured on stressed and unstressed plants.

Determination of Yield Stability
Yield stability index (YSI) was determined as the slope of the regression between the yield of individual genotypes and the mean yield of all the genotypes recorded at various moisture levels of the line source (Field Exp. 1). The stability index, if <1, indicates higher yield stability, whereas it indicates poor yield stability if it is >1 (Eberhart and Russell, 1966). Stability in grain yield was also estimated for each genotype using the drought susceptibility index (DSI) derived from the yield difference between stress and nonstress environments (Fischer and Maurer, 1978). The DSI estimates for each genotype the rate of change in yield between the stress and nonstress environments relative to the mean change for all genotypes, as follows:

where Y is yield under stress, Yp is yield without stress, and X and Xp represent average yield over all varieties under stress and nonstress conditions, respectively. Line Source Position 1 and 5 were considered as nonstress and stress condition, respectively. Values of DSI higher than 1 denote low drought susceptibility (or higher yield stability) while the values lower than 1 indicate high drought susceptibility (or poor yield stability).

Estimation of Osmotic Adjustment Heritability
Genetic variance of OA (as well as of OP100 and TP) was computed using pooled analysis of variance across the six greenhouse experiments, and heritability of these traits was determined by the ratio of genetic to total variance (Allard, 1960). The broad-sense heritability (h2) for mean values over the environments was calculated as:

where {sigma}g2 stands for genetic variance while {sigma}ge2 and {sigma}{epsilon}2 stand for variance due to genotype x environment interaction and residual error, respectively. E and R are the number of environments and replicates, respectively. Each greenhouse experiment was considered as a unique environment.

Statistical Analyses
Means and standard errors were calculated according to the standard statistical procedure laid down by Gomez and Gomez (1984). Statistical analyses for the line source experiment were performed assuming that the experiment was a split block design with irrigation levels arranged systematically in each replicate along the gradient of applied water and genotypes randomized in each replicate. Analysis of variance (ANOVA) was performed to examine treatment effects in Field Exp. 2. Simple linear regression and correlation analyses were employed to express the relationship among variables of interest.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Variation among Genotypes and Heritability of Osmotic Adjustment
As computed by the pooled analysis of variance across the six greenhouse experiments, genetic variation among the bread wheat genotypes was 0.149 for OA, 0.113 for OP100, and 0.170 for TP, whereas the heritability of OA, OP100, and TP was 0.74, 0.73, and 0.79, respectively (Fig. 2) . As for the mean phenotypic variation across the eight bread wheat genotypes, OA ranged from 0.39 to 0.57 MPa while OP100 and TP ranged from –1.72 to –2.38 and 0.14 to 0.62 MPa, respectively, at a WP ranging from –1.56 to –2.44 MPa. Considering the mean across the six greenhouse experiments, OA ranged from 0.37 to 0.59 MPa while OP100 and TP ranged from –1.80 to –2.11 and 0.33 to 0.52 MPa, respectively, at a WP of –1.90 to –2.17 MPa (Table 5). The absolute values of OA, OP100, and TP, measured in the greenhouse experiments, ranged from 0.31 to 0.86 MPa, –1.62 to –2.95 MPa, and 0.27 to 0.76 MPa, respectively, whereas the OA and OP100, measured in Field Exp. 2, ranged from 0.60 to 0.99 MPa and –2.28 to –2.72 MPa, respectively (data not shown).



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2. Genetic variance and heritability of osmotic adjustment (OA), osmotic potential at full turgor (OP100), and turgor potential (TP) computed by the pooled analysis of variance across the six greenhouse experiments.

 

View this table:
[in this window]
[in a new window]
  		Table 5. Mean values along with standard error (±SE) of osmotic adjustment (OA), water potential (WP), osmotic potential at full turgor (OP100), and turgor potential (TP) across the six greenhouse experiments and eight bread wheat genotypes.

 
Association of Osmotic Adjustment with Grain Yield
Regression analysis was performed between OA measured in the six greenhouse experiments and the grain yield recorded at different moisture levels of the line source experiment (Field Exp. 1). Generally, the value of correlation coefficient increased with the increase of water deficit, i.e., the greater the water deficit, the stronger the association between OA and grain yield (Table 6). There was similar correlation of OP100 and TP measured in the greenhouse and grain yield obtained at various moisture levels of the line source. In several cases, the correlation coefficient values turned positively significant (P < 0.10 or P < 0.05) at fifth and/or fourth moisture level. The correlation between osmotic parameters (OA, OP100, and TP) and grain yield at fifth moisture level was weak (P < 0.10) in several cases, presumably due to less degrees of freedom (df) because only eight genotypes were included in the correlation analysis (n = 8). Such an inference was drawn because the correlation of OA, OP100, and TP was generally significant (P < 0.05 or P < 0.01) with grain yield attained at fifth moisture level of the line source when 15 genotypes of Exp. 4–6 were used in the correlation analysis (n = 15) (Table 7). Nonetheless, there was highly significant (P < 0.001) correlation between mean values of OA across the six greenhouse experiment and those of grain yield at fifth moisture level of the line source (Table 6) even with eight genotypes (n = 8). Similarly, there appeared significant (P < 0.05) positive correlation when the mean values of OP100 and TP across the six greenhouse experiments were regressed against those of grain yield attained at the fifth moisture level of the line source (Table 6). Moreover, there was quite strong correlation (P < 0.01) between mean values of OP100 and grain yield, when 15 genotypes were included in the correlation analysis, using Exp. 4–6 (Table 7). Of the six greenhouse experiments, Exp. 1 showed the most strong (P < 0.01) correlation between OA (measured in the greenhouse) and grain yield recorded at fifth moisture level of the line source even with limited number of observations (n = 8) (Table 6).


View this table:
[in this window]
[in a new window]
  		Table 6. Simple linear correlation of osmotic traits with grain yield attained at five moisture levels of the line source and yield stability indices (n = 8).

 

View this table:
[in this window]
[in a new window]
  		Table 7. Simple linear correlation between osmotic traits and grain yield at five moisture levels of the line source (Field Exp. 1) and yield stability indices (n = 15).

 
There was a significant (P < 0.05) positive correlation between OA and grain yield under simulated water deficit conditions in the field (Tables 8 and 9). Osmotic adjustment was correlated with grain yield, biomass, and number of grains per square meter under severe water deficit condition (total water applied including rainfall = 202 mm) of Field Exp. 2 (Table 8). Osmotic potential at full turgor and grain yield were also correlated (P < 0.10). Moreover, there was significant (P < 0.05) positive correlation between OA measured in Field Exp. 2 and grain yield obtained at two moisture levels of the double line source experiment (Field Exp. 3) (Table 9). The grain yield decreased with the increase in water stress as per the standard line source moisture gradient on one side of LS1. However, correlation coefficient values due to regression between OA and grain yield increased with the increase in water stress. That is, the greater the water deficit, the stronger the association between OA and grain yield, with the correlation coefficient value turning significant (P < 0.05) at fourth moisture level. There was similar correlation of OA and OP100 (Field Exp. 2) with grain yield (Field Exp. 3) as a result of preanthesis soil moisture gradient in between the two line sources (LS1 and LS2). The uniform preanthesis stress, followed by a gradient of postanthesis stress on the other side of LS2, could not result in any significant (P < 0.05) correlation between OA (Field Exp. 2) and grain yield (Field Exp. 3); however, there was significant (P < 0.05) correlation between OP100 and grain yield at second moisture level (Table 9).


View this table:
[in this window]
[in a new window]
  		Table 8. Total water applied including rainfall, grain yield, and yield parameters in association with osmotic adjustment (OA) and osmotic potential at full turgor (OP100) (Field Exp. 2).{dagger}

 

View this table:
[in this window]
[in a new window]
  		Table 9. Total water applied including rainfall, grain yield, and the linear relationship of osmotic adjustment (OA) and osmotic potential at full turgor (OP100), determined in Field Exp. 2, with grain yield attained at five moisture levels of the double line source experiment (Field Exp. 3).{dagger}

 
Association of Osmotic Adjustment with Yield Stability
There was consistent negative correlation between OA (greenhouse experiments) and YSI (Field Exp. 1) (Tables 6 and 7). However, the correlation was statistically significant (P < 0.05) only in Exp. 4 (Table 6) and 5 (Table 7). The correlation between OP100 and YSI was also significant (P < 0.05 or P < 0.10) in Exp. 4 (Table 6) and 5 (Table 7). Mean values of OA and OP100 were significantly correlated (P < 0.05) with YSI when 15 genotypes were used in the correlation analysis (Exp. 4–6) (Table 7). Turgor potential was significantly correlated (P < 0.05 or P < 0.01) with YSI in Exp. 1–3 (Table 6) and 5 (Table 7). Besides, there was significant (P < 0.05 or P < 0.001) correlation between mean TP across the greenhouse experiment and YSI (Tables 6 and 7).

Osmotic adjustment was positively correlated (P < 0.05 or P < 0.01) with DSI in the case of Exp. 1, 4 (Table 6), and 5 (Table 7). Besides, there was significant (P < 0.05, P < 0.01, or P < 0.001) correlation between OP100 and DSI regarding Exp. 3, 4 (Table 6), and 5 (Table 7). The correlation between mean OP100 and DSI was only significant in the case of Exp. 4–6 (n = 15). The correlation between TP and DSI was significant (P < 0.05, P < 0.01, or P < 0.001) for Exp. 1–3 (Table 6) and 5 (Table 7). The mean TP across the greenhouse experiments was also significantly correlated (P < 0.05 or P < 0.001) with DSI (Tables 6 and 7).

There was a significant (P < 0.05) negative correlation between YSI and grain yield and a significant (P < 0.05) positive correlation between DSI and grain yield attained at fifth moisture level of the line source (Field Exp. 1) (Fig. 3) . That is, the lower the YSI (the higher the yield stability) or the higher the DSI (the lower the drought susceptibility), the higher the grain yield under water deficit.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 3. (A and B) Relationship of yield stability index (YSI) and drought susceptibility index (DSI) with grain yield attained at fifth moisture level (T5) of the line source, and (C) the relationship between YSI and DSI (n = 8) (Field Exp. 1).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Variation among Genotypes and Heritability of Osmotic Adjustment
The present investigation reveals considerable genetic variation and heritability of OA, OP100, and TP as computed by pooled analysis of variance across the six greenhouse experiments (Fig. 2). Besides, this study also confirms the results of other workers (Morgan, 1977, 1995, 1999; Morgan et al., 1986; Blum et al., 1999) regarding substantial phenotypic variability of OA (0.31 to 0.86 MPa) among bread wheat genotypes under water deficit environments (Table 5). The induction of OA under simulated drought conditions in the field was comparatively greater (0.60 to 0.99 MPa) due to the expected slow development of water stress in the field compared with that in the greenhouse (Turner and Jones, 1980; Morgan, 1984; Flower and Ludlow, 1986). The wide range of variability among genotypes also indicates the suitability of OA as selection tool for breeders under water deficit environments. This study also showed similar results regarding OP100 and TP (Table 5).

Association of Osmotic Adjustment with Grain Yield
The correlation between OA measured in the greenhouse experiments and the grain yield attained at various moisture levels of the line source (Field Exp. 1) indicates that the association of OA with the grain yield under water deficit depends on the intensity of water deficit (Tables 6 and 7). That is, the greater the water deficit, the stronger the association between OA and grain yield, as earlier indicated by Morgan (1983)(1984). When the mean values of OA (as well as of OP100 and TP) across the greenhouse experiment were regressed against grain yield obtained at various moisture levels of the line source, the correlation coefficient value was highly significant (P < 0.01 or P < 0.001) at the fifth moisture level (Tables 6 and 7). This is because averaging the values across the greenhouse experiments could minimize the experimental error associated with different greenhouse experiments. The correlation between OA measured in the greenhouse and grain yield recorded in Field Exp. 1 (Tables 6 and 7) also coincided with the correlation between OA and grain yield measured in Field Exp. 2 (Table 8). It also reflected the correlation between OA measured in Field Exp. 2 and grain yield recorded at various moisture levels of Field Exp. 3 (Table 9). These results are expected as the induction of OA takes place only in stressed condition as per the intensity of water stress (Morgan, 1984). The gene for OA induction has already been recognized in wheat and rice (Oryza sativa L.) and is expressed in stressed condition only. The differences in OA are conditioned by alternative alleles of OA gene at a single major locus (Morgan, 1983; Morgan et al., 1991; Morgan and Tan, 1996; Lilley et al., 1996; Zhang et al., 1999; Zheng et al., 2000).

The present results demonstrate that OP100 and TP, measured on stressed plants in the greenhouse, also behaved like OA (Tables 6 and 7), which is obvious as OA occurs due to accumulation of solutes in the plants during water deficit, as a result of which OP is decreased and, thereby, maintenance of turgor takes place at lowered WPs (Turner and Jones, 1980; Morgan, 1984; Ludlow and Muchow, 1990). This also finds support by a positive association (P < 0.05) recorded in this study between mean values of OA and those of OP100 and TP across the six greenhouse experiments (Fig. 4) . In fact, a highly significant positive association (P < 0.001) between OA and OP100 has been reported by Subbarao et al. (2000a) in the case of pigeonpea under water deficit.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Relationships of osmotic adjustment with (A) osmotic potential at full turgor and (B) turgor potential. All data points represent mean values across the six greenhouse experiments.

 
In addition to a significant (P < 0.05) correlation between OA and grain yield recorded at severe water stress level, OA was also correlated positively (P < 0.05) with biomass as well as grains per square meter in Field Exp. 2 (Table 8). These results corroborate the findings of Blum et al. (1999) regarding the positive correlation of OA with biomass and grain yield of wheat and those of Wilson and Eastin (1982), Ludlow et al. (1990), Santamaria et al. (1990), and Tangpremsri et al. (1995) regarding the positive association of OA and grain number in the case of grain sorghum. Moinuddin and Renu Khanna-Chopra (2004) also reported this in case of chickpea (Cicer arietinum L.).

Among the six greenhouse experiments, Exp. 1 appears to be the best protocol to measure OA for screening the improved bread wheat genotypes for water deficit environment because it reflected a comparatively greater capacity of OA (Table 5) and resulted in a quite significant (P < 0.01) correlation between OA and grain yield at fifth moisture level of the line source (Table 6), even with the limited number of observations (n = 8). In fact, the plants of Exp. 1 experienced the lowest maximum (28.7°C) as well as mean (20.25°C) temperature (Table 2) during the soil drying cycle, which could result in a comparatively slow drying of the soil and, in turn, lead to a greater induction of OA (Morgan, 1984). Expectedly, the OA, induced by the simulated drought condition at the peak reproductive stages (from anthesis to grain filling) in Exp. 1, could have resulted in a comparatively higher seed set (Wright et al., 1983b; Santamaria et al., 1990; Moinuddin and Renu Khanna-Chopra, 2004) and a better translocation of preanthesis photoassimilates to the developing grains (Wright et al., 1983a; Ludlow et al., 1990; Subbarao et al., 2000b), thus resulting in a quite strong (P < 0.01) correlation between OA and grain yield at the highest water deficit level of the line source. Besides, there was a significant (P < 0.05) correlation between OA and DSI, indicating the important role of OA in yield stability under water deficit condition of Exp. 1 (Table 6).

Osmotic Adjustment and Yield Stability
Consistent negative correlation of OA with YSI and similar positive correlation of OA with DSI indicates the important role of OA in maintaining the yield stability under water deficit (Tables 6 and 7). Thus, it might be inferred that OA could stabilize the yield under water deficit by protecting the crop from drought susceptibility as a result of maintenance of turgor. This finds support from a significant (P < 0.05 or P < 0.001) correlation of mean TP with YSI and DSI (Tables 6 and 7) and that recorded between the two indices (Fig. 3). In fact, Ludlow et al. (1990) estimated that an OA of 0.6 MPa could prevent any reduction in yield of sorghum due to a postanthesis water stress of the degree of –1.7 to –1.8 MPa (WP). The contribution of OA to yield stability under drought environments has also been recorded in case of wheat and barley (Hordeum vulgare L.) (Morgan et al., 1986; Blum and Pnuel, 1990; Teulat et al., 1997). Further, Renu Khanna-Chopra et al. (1994) and Renu Khanna-Chopra (1999) have reported significant (P < 0.05) negative correlation between OA and YSI in the case of wheat genotypes.

In conclusion, the present investigation reveals substantial differences in OA among the bread wheat genotypes. A high positive correlation (P < 0.001) between OA of greenhouse-stressed plants and the grain yield recorded in the field under water deficit indicates that improved bread wheat cultivars could be successfully screened for drought environments in the greenhouse on the basis of OA. The methodology employed in Greenhouse Exp. 1 could be successfully used for screening bread wheat genotypes in this regard. Besides, a significant (P < 0.05) correlation between OA and grain yield under simulated water deficit condition in the field further advocates that OA could be regarded as a selection criterion for screening improved bread wheat genotypes for water deficit environments. Moreover, this study also indicates that OA is a heritable character and plays an important role in maintaining yield stability under water deficit. The present study also reveals similar results regarding OP100 and TP, both of which could also be used for screening improved bread wheat genotypes for water deficit environments, using greenhouse methodology.


   ACKNOWLEDGMENTS
 
The facilities provided by CIMMYT regarding this investigation are highly acknowledged. The senior author is grateful to governments of Mexico and India for providing the financial assistance. Thanks are also due to Dr. Rajaram (Director, Wheat Program, CIMMYT, Mexico) in connection with availability of various bread wheat genotypes used in this study.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Moinuddin,
Right arrow Articles by Reynolds, M. P.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Moinuddin,
Right arrow Articles by Reynolds, M. P.
Agricola
Right arrow Articles by Moinuddin,
Right arrow Articles by Reynolds, M. P.
Related Collections
Right arrow Water Stress
Right arrow Crop Physiology & Metabolism

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Crop Science Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome