HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Desclaux, D. Articles by Roumet, P. Search for Related Content PubMed Articles by Desclaux, D. Articles by Roumet, P. Agricola Articles by Desclaux, D. Articles by Roumet, P.

CROP ECOLOGY, MANAGEMENT & QUALITY

Identification of Soybean Plant Characteristics That Indicate the Timing of Drought Stress

^a INRA, Station de Génétique et d'Amélioration des plantes, 34130 Mauguio, France
^b Cantho Univ., Dep. of Genetics and Plant Breeding, 30 April Street, Cantho, Viet Nam

desclaux{at}ensam.inra.fr

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 
Low-cost, phenotype-based techniques are needed to help crop breeders interpret genotype x environment interactions. Our objective was to determine how drought stress imposed on soybean [Glycine max (L.) Merr.] at various growth stages affected selected plant characteristics. Determinate (Spot) and indeterminate (Weber) soybean cultivars were grown under greenhouse conditions and subjected to two levels of drought stress (30 and 50% plant-available water) during vegetative (V4–R1), flowering (R1–R3), pod lengthening (R3–R5), or seed-filling (R5) stages. Mean internode length was the most drought sensitive factor during the vegetative and flowering stages. Significant height differences for the determinate cultivar accurately differentiated between stress periods, with shorter plants being associated with vegetative stress. The number of pods per vegetative dry matter unit was significantly affected by stress during pod lengthening. Early stress during seed fill reduced the number of seeds per pod, whereas late stress (after the abortion limit stage) decreased seed weight. These results suggest that precise periods of drought stress can be determined by measuring a posteriori several morphological factors and yield components of determinate and indeterminate soybean genotypes at physiological maturity.

Abbreviations: i.n.l, internode length • i.s.w, individual seed weight • FS, drought stress during flowering stage • PAW, plant-available water • PLS, drought stress during pod lengthening stage • SFS, drought stress during seed-filling stage • vdm, vegetative dry matter • VS, drought stress during vegetative stage

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 
WATER is the main factor limiting soybean production in France (Blanchet et al., 1988; Merrien, 1994). For the Maturity Group I and II adapted to the largest French production area, plant genotype x environment interactions are important. To interpret them, crop breeders need to characterize each yield trial location. Agronomic tools such as soil analyses, weather data, tensiometer or neutron probe records, leaf water potential data, or other drought stress measurements are useful, but experimental yield trials are generally large scale and conducted at locations where these tools may not be available. Moreover, it is often difficult to find, among the great number of collected data, the period when water was limiting and if a water deficit occurred, to deduce if it had an impact on crop development. To avoid these problems and help to focus the research among all climatic and edaphic data, Meynard and David (1992) proposed for wheat (Triticum aestivum L.) an environmental diagnosis based on cultivar observations. Yield components are determined at specific growth stages and each is subject to the environmental conditions that prevail during this period. Analysis of each component value could thus provide a post-harvest assessment of when environmental stress occurred and how severe it was. The main interest of the method is to avoid wide ranging research on limiting factors from a common climatic database by focusing on specific periods, and to provide complementary observations. Indeed, when describing an environment, the important point is to know whether plants have suffered from a water deficit rather than simply that this deficit occurred.

For wheat, Meynard and David (1992) focused observations on two yield components that are temporally separated by anthesis, but for indeterminate soybean, there is an overlap during which seed number and seed weight are determined. Therefore, to use soybean plant components to diagnose the presence of water deficits, grain yield must be broken down into more components. To choose the most relevant components, specific effects on drought stress on each must be determined.

Most studies quantifying the impact of drought on soybean have focused mainly on yield components of different genotypes (Sionit and Kramer, 1977; Momen et al., 1979; Korte et al., 1983; Meckel et al., 1984; Specht et al., 1986; Andriani et al., 1991; Smiciklas et al., 1992; Vieira et al., 1992). However, a few have reported the effects of different drought intensities on determinate and indeterminate soybean during various growth stages. Desclaux and Roumet (1996) and others have reported that drought stress seemed to trigger an early switch from vegetative to reproductive development. Depending on when stress occurred, different phenological changes were observed. The appearance of nodes initiated during stress was delayed, resulting in a smaller number of nodes produced, whereas flower and pod appearance was hastened. Seed-filling and final seed abortion stages began earlier and overall each reproductive phase was shorter under stress.

Our objective in this report is to establish specific relationships between when drought stress occurred and its effects on soybean morphological variables and yield component values. We will try to isolate the variables that best discriminate among stress periods and among stress intensities.

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 
Plant Material and Culture Conditions
A greenhouse study was conducted at the Station de Génétique et d'Amélioration des plantes, INRA, Montpellier, France (43°34' N, 3°57' E). Indeterminate (Weber) and determinate (Spot) Maturity Group I cultivars were inoculated with Bradyrhizobium japonicum (G49 strain) and sown in polyvinylchloride pots (155-mm diameter, 300-mm height) filled with 8 kg of a sandy loam soil (1:4 ratio) on 16 April 1993 and 2 May 1994. One week after sowing, the pots were thinned to one seedling each, providing a plant density of 36 plants m^-2. Maximum day temperatures were maintained at about 26°C, night temperatures averaged 19°C, and relative humidity was regulated at 85%. No artificial light was used.

Treatments and Experimental Design
For each cultivar, 200 pots were separated into five groups. One group of 40 was watered daily as the control treatment. The other four groups were subjected to drought stress during one of the following developmental stages: (i) vegetative stress (VS) applied when the plants had four fully expanded leaves (20 d old) and ended when flowering began (R1: Fehr and Caviness, 1977); (ii) flowering stress (FS) imposed from the beginning of the flowering period (R1) to first pod appearance (R3); (iii) pod lengthening stress (PLS) applied when the first pod appeared (R3) and ended at the beginning of pod filling (R5); and (iv) seed-filling stress (SFS) for which water was withheld for 15 d from R5. Except when subjected to drought stress, plants were watered daily with nutrient solution (Matsumoto et al., 1975) to maintain soil water at pot capacity.

For each drought treatment, water was withheld until the pots reached stress levels of either 50% or 30% of plant-available water (PAW). The upper PAW limit was determined by weighing soil from five pots 2 d after they were watered. The lower PAW level was determined by weighing pots in which plants were allowed to die after transpiring all available water. Stress levels for this study are denoted by the number 1 (50%) or 2 (30%) indexing the stress period (i.e., FS₁, FS₂...). In all cases, water was withheld for 4 to 5 d to achieve the various stress intensities. After that time, each pot was weighed daily and water was added if necessary to maintain the stress level. In both years, each stress period lasted about 15 d.

In addition to daily pot weighing, leaf water potential (Schölander et al., 1965) and soil water potential at 0.2-m depth (Tensiometric System DTE 1000, Nardeux, St Avertin, France) were also monitored. Leaf water potential was measured with a pressure chamber (Model 1000, PMS Instrument Co, Corvallis, OR) on the most recently fully expanded leaf, at 1200 h on four randomly selected plants, three times weekly for each treatment. No measurements were made under VS.

Measurements
At maturity, morphological variables and yield components were measured on five plants for each treatment within both replicates. Mainstem height was measured from the collar to the top, the number of nodes on the main stem was counted using the cotyledonary node as Node 1, and the number of branches, their length, and number of nodes were also measured. The number of pods and seeds carried by the mainstem and by branches were counted. Samples were dried for 72 h at 80°C. Vegetative and reproductive dry weights were estimated at a water content of 0 g kg^-1. Time was expressed in degree days cumulated from sowing with a base temperature of 0°C.

Statistical Analysis
The experimental design consisted of a randomized complete block with two replicates for each of the two cultivars, five growth stages, and two stress levels.

Data were analyzed by analysis of variance for each genotype for each treatment within years and across years. Appropriate F-tests were used to assess significance and means were compared by the Newman-Keul's multiple range test (SAS Institute, 1989). To avoid excessive repetition, all morphological and yield components data will be presented hereafter for the indeterminate cultivar Weber. The determinate cultivar Spot, differing essentially in its vegetative pattern, will be used to specify the impact of drought stress before flowering, and only the data concerning morphological traits will be presented.

	Results

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 
Stress Timing and Intensity
The drought stress treatments were imposed at different growth stages characterized by the Fehr and Caviness (1977) scale, which is based on observations of the four upper mainstem nodes. For the indeterminate genotype, the overlap of stages on a node basis is shown in Fig. 1 . The reader should note that during a treatment such as PLS (pod lengthening stress), flowers are still present on the mainstem upper nodes, whereas seeds are already formed at the lower nodes.

View larger version (26K): [in this window] [in a new window]   Fig. 1 Appearance of reproductive organs on the mainstem of Weber soybean as a function of cumulative degree-days (base 0°C), under well-watered conditions (Flo: first flower, Pod: first pod, Seed: first seed, Fin.S.A: final stage in seed abortion). VS, FS, PLS, and SFS represent the following stress periods: vegetative, flowering, pod lengthening, seed filling, respectively

View this table: [in this window] [in a new window]   Table 1 Soil water content (g kg-1) during each treatment in relation to the maximum quantity of water that pots could retain

View larger version (26K): [in this window] [in a new window]   Fig. 2 Leaf water potential of Weber soybean cultivated under well-watered conditions (control) or under drought stress treatments differing by their intensity levels (1 = moderate, 2 = severe) and timing: flowering (FS), pod lengthening (PLS), seed filling (SFS), during the 2-yr experimental period (1993 and 1994)

View larger version (25K): [in this window] [in a new window]   Fig. 3 Notation scale for Weber nodes according to the stage of axillary leaf development in the control treatment. The two horizontal arrows indicate the period of applied treatments (FS or PLS)

View larger version (22K): [in this window] [in a new window]   Fig. 4 Mainstem internode length of Weber soybean cultivated under well-watered conditions (control) or under drought stress during flowering (FS) or during pod lengthening (PLS). Each point represents a mean of six measurements and stars indicate significant differences from the control. The two vertical arrows indicate node ranks that were emitted during treatments (FS or PLS)

Number of Pods
Each stress treatment decreased the number of pods in different ways. Early stress (VS or FS) reduced the number of flowers that were produced (i.e., FS plants had only half as many flowers as the control). This reduction was closely correlated with lower dry matter produced under these treatments. Indeed, early stress resulted in only 45% of the vegetative dry matter produced by control plants. Stress during PLS drastically increased the pod abscission rate (>95%) compared with the control.

Measured at maturity, the number of pods was not a variable that could help to discriminate among stress periods because it was too dependent on the values of previous components and especially of the vegetative dry matter. Therefore, we examined the pod number per unit dry matter ratio (Fig. 5) . This ratio was significantly different from the control only for plants subjected to drought stress during PLS. The decrease was greater in 1994 than 1993 because, as shown in Table 1 and Fig. 2, the intensity of stress was higher in 1994. The lack of differences in this ratio between early stress (VS, FS) and control plants confirmed that the low number of pods, observed on Weber plants submitted to VS or FS, was essentially due to a vegetative dry matter deficiency.

View larger version (26K): [in this window] [in a new window]   Fig. 5 Number of pods produced per gram of vegetative dry matter, as a percentage of Weber soybean control, for the two experimental years. Each point represents a mean of 10 plants. Stars indicate significant differences from the control at the 0.05 significance level for all points accompanied by vertical line. A lack of a star means no difference from the control

View larger version (16K): [in this window] [in a new window]   Fig. 6 Pattern for the number of seeds per pod in Weber soybean plants cultivated under well-watered conditions (control) and under drought stress during seed filling (SFS1, SFS2) differing by intensity level (1 = moderate, 2 = severe) in 1994. Each point is the mean of 10 plants, crosses indicate significant differences from the control at the 0.05 significance level

View larger version (20K): [in this window] [in a new window]   Fig. 7 Patterns for the individual seed weight (mg) of Weber soybean plants cultivated under well-watered conditions (Control), and under drought stress applied during the vegetative period (VS), flowering (FS), pod lengthening (PLS) and seed-filling period (SFS: intensity level 1 = moderate, 2 = severe) in 1994

	Discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 
Average internode length, pods per unit dry matter, seeds per pod, and seed weight were plotted as a percentage of the control for each stress treatment (Fig. 8) to determine which might be useful for identifying the timing and intensity of drought stress in soybean.

View larger version (26K): [in this window] [in a new window]   Fig. 8 Yield elaboration course for Weber soybean subjected to drought stress during the vegetative period (a), flowering (b), pod lengthening (c), and seed filling (d). Internode length (i.n.l), number of pods per vegetative dry matter (pod/vdm), number of seed per pod (seed/pod) and individual seed weight (i.s.w) are plotted on the x-axis according to the end of elaboration period and on the y-axis as a percentage of control. Each point is the mean of 10 plants. Stars indicate significant differences from the control at the 0.05 significance level. Vertical lines accompanying stars indicate that the differences are significant for all points. A lack of a star means no difference from the control

Effect of Drought Stress Applied During Flowering
Internode length was reduced by stress at flowering (FS). For the indeterminate cultivar (Weber), the mainstem height at flowering represented only 50% of the final height. Using the determinate cultivar (Spot) also enabled us to differentiate between stress at vegetative (VS) and flowering (FS) (Table 2). Drought stress during flowering enhanced dry matter efficiency for pod production (in terms of number). This agrees with results obtained by Deumier and Ney (1989) for pea (Pisum sativum L.) and indicates that low dry matter accumulation was the main cause for low pod and seed number. Low pod number was essentially linked to the decreased number of flowers produced and could be due to seed abortion at an early stage as Westgate and Peterson (1993) proposed.

An increased number of seeds per pod for plants subjected to low intensity drought stress was previously reported by Andriani et al. (1991), who imposed a stress equal to 50% PAW and measured a leaf water potential of -1.3 MPa (Scopel, 1993). That level of stress corresponded to our FS₁ treatment in 1993. With greater drought stress (<40% PAW), however, there were no differences when compared with control plants.

The individual seed weight of plants cultivated under FS₁ in 1994 was significantly higher than the control and seed abortion was lower for the two FS treatments. This compensation phenomenon for the low number of seeds per pod was previously reported by Mingeau (1975), Vidal et al. (1981), and Andriani et al. (1991).

Effect of Drought Stress Applied During Pod Lengthening
Drought stress imposed during pod lengthening had the greatest effect on the number of pods produced per unit of dry matter (Fig. 8c), a response already noted by several authors (Momen et al., 1979; Kadhem et al., 1985; Son et al., 1996). The correlation coefficient between the number of pods on the mainstem and the yield of the plant was 0.70. The decreased number of pods was observed on all nodes but was essentially due to abortion of the youngest pod. The high pod abortion and good vegetative development of plants subjected to the PLS treatment facilitated good seed filling in the remaining pods, especially in 1994 when the stress intensity was much more intense than in 1993. Drought stress applied before the abortion limit stage thus favored the integrity (weight and quality) of the first seeds to the detriment of their number.

Effect of Drought Stress Applied During Seed Filling
The phenotype of SFS plants was marked by the presence of many flat pods (with no seeds) on the upper mainstem nodes. These flat pods were the youngest that had not yet reached the abortion limit stage when the stress was imposed. On the first reproductive nodes, drought stress decreased the seed weight and led to some seed abortion in the pod. Individual seed weight was more affected by high intensity stress (SFS₂), correlating significantly (P < 0.01) with PAW. The positive correlation (r² = 0.7) between yield and the length of the seed-fill period is consistent with the studies of Bradford (1994) and Morandi et al. (1994). The decreased seed weight observed could therefore be the result of a shorter filling period (almost 7 d in our experiments).

	Conclusions

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 
This greenhouse study of mature plant phenotype highlighted significant relationships between the drought stress period and plant characteristics. These characteristics are expressed as ratios between variables that occur consecutively during soybean development. A component value indicating plant response depends not only on environmental conditions that prevail during its elaboration but also on the values of previous components.

The ratio height/node number (internode length) was useful for diagnosing early stress. Early stress reduced vegetative biomass production and decreased the average internode length without altering the efficiency of dry matter for producing and filling pods and seeds. To differentiate stress before flowering from that occurring during anthesis, analysis of the height of a determinate cultivar was useful.

The number of pods per unit dry matter weight was an effective indicator of water limitation during pod lengthening, especially for Weber. Spot reacted less for this character. Plants subjected to drought stress after flowering but before the abortion limit stage, favored the bearing of the first seeds to the detriment of their number.

Our results indicate that the number of seeds per pod can be used as an estimator of seed abscission rate, assuming that, for a genotype, the average ovule number doesn't fluctuate and depends on water conditions before abortion limit stage. Severe stress can induce an important reduction of this ratio, since in our experiment it was reduced 35% in the stressed treatment relative to the control. Stress applied after the abortion limit stage decreased seed fill duration and seed weight. The response of these four plant characteristics to drought can assist a researcher in identifying the period during development that the stress may have occurred.

Received for publication November 17, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results Discussion Conclusions REFERENCES

 

• Andriani J.M., Andrade F.H., Suero E.E., Dardanelli J.L. Water deficits during reproductive growth of soybeans. I. Their effects on dry matter accumulation,and its components. Agronomie 1991;11:77-746.
• Blanchet R., Desclaux D., Gelfi N., Quinones H. Simulation des principales contraintes agronomiques limitant le rendement du soja dans différentes situations françaises. C.R. Acad. Agric. Fr. 1988;74:11-19.
• Bradford K.J. Water stress and the water relations of seed development: A critical review. Crop Sci. 1994;34:1-11.
• Desclaux D., Roumet P. Impact of drought stress on the phenology of two soybean cultivars. Field Crops Res. 1996;46(1–3):61-70.
• Deumier, J.M., and B. Ney. 1989. Irrigation du pois de printemps. Atout Pois, ITCF-UNIP ed., p123–130.
• Elmore R.W., Eisenhauer D.E., Specht J.E., Williams J.H. Soybean yield and yield component response to limited capacity sprinkler irrigation systems. J. Prod. Agric. 1988;1:196-201.
• Giovanardi, R., P. Ceccon, T. Fasan. 1990. Agronomic and physiological response of soybeans (Glycine max (L.) Merr.) to water stress at different development stages. II. Yield, yield components and quality aspects. Rivista di Agronomia 24:2–3, 218–227.
• Fehr W.R., Caviness C.E. Stages of soybean development. Iowa State Univ, Ames: Special Rep 80. Coop. Ext. Ser. Agric. and Home Econ. Exp. Stn, 1977.
• Kadhem F.A., Specht J.E., Williams J.H. Soybean irrigation serially timed during stages R1 to R6. II. Yield component responses. Agron. J. 1985;77:299-304.[Abstract/Free Full Text]
• Korte L.L., Specht J.E., Williams J.H., Sorensen R.C. Irrigation of soybean genotypes during reproductive ontogeny. II. Yield component responses. Crop Sci. 1983;23:528-533.[Abstract/Free Full Text]
• Matsumoto T., Yamamoto Y., Yatazawa M. Role of root nodules in the nitrogen nutrition of soybeans. I. Fluctuation of allantoin and some other plant constituents in the growing period. J. Sci. Soil Manure 1975;46:471-477.
• Meckel L., Egli D.B., Phillips R.E., Radcliffe D., Leggett J.E. Effect of moisture stress on seed growth in soybeans. Agron. J. 1984;76:647-650.[Abstract/Free Full Text]
• Merrien, A. 1994. Soybean water management under French conditions. p. 36. In World Soybean Research Conference, V, Chiang Mai, Thailand. 20–26 Feb. 1994. Funny Publishing Limited Partnership, Bangkok.
• Meynard J.M., David G. Diagnostic de l'élaboration du rendement des cultures. Cahiers Agricultures 1992;1:9-19.
• Mingeau M. Etude de la sensibilité du soja à la sécheresse. Info. Tech. CETIOM 1975;47:1-14.
• Momen N.N., Carlson R.E., Shaw R.H., Arjmand O. Moisture stress effects on the yield components of two soybean cultivars. Agron. J. 1979;71:86-90.[Abstract/Free Full Text]
• Morandi, E.N., M.L. Boderro, R.A. Martigone, and A. Quijano. 1994. Sowing date and irrigation effects on soybean dry matter partitioning and yield in the southern Santa Fe area of Argentina. p. 128. In World Soybean Research Conference, V, Chiang Mai, Thailand. 20–26 Feb. 1994. Funny Publishing Limited Partnership, Bangkok.
• SAS Institute. SAS/STAT guide for personal computers, version 6 ed. Cary, NC: SAS Institue, 1989.
• Schölander P.F., Hammel H.T., Bradstreet E.D., Hemmingsen E.A. Sap pressure in vascular plants. Negative hydrostatic pressure can be measured in plants. Science 1965;148:339-346.[Abstract/Free Full Text]
• Scopel, E. 1993. Croissance et développement de deux génotypes isogéniques, déterminé et indéterminé de soja. Effets d'un déficit hydrique durant la floraison. Thèse ENSA. Montpellier, France.
• Sionit K., Kramer P.J. Effect of water stress during different stages of growth of soybean. Agron. J. 1977;69:274-278.[Abstract/Free Full Text]
• Smiciklas K.D., Mullen R.E., Carlson R.E., Knapp A.D. Soybean seed quality response to drought stress and pod position. Agron. J. 1992;84:166-170.[Abstract/Free Full Text]
• Son C.K., Park S.G., Kim C.Y., Choi B.S. Effects of water stress on growth characteristics and grain yield of soybean. RDA J. Agric. Sci. Upland Industrial Crops 1996;38(2):123-128.
• Specht J.E., Williams J.H., Weidenbenner C.J. Differential responses of soybean genotypes subjected to a seasonal soil water gradient. Crop Sci. 1986;26:922-930.[Abstract/Free Full Text]
• Vidal A., Arnaudo D., Arnoux M. La résistance à la sécheresse du soja. I-influence d'un déficit hydrique sur la croissance et la production. Agronomie 1981;1(4):295-302.
• Vieira R.D., Tekrony D.M., Egli D.B. Effect of drought and defoliation stress in the field on soybean seed germination and vigor. Crop Sci. 1992;32:471-475.[Abstract/Free Full Text]
• Westgate M.E., Peterson C.M. Flower and pod development in water-deficient soybeans. J. Exp. Bot. 1993;44:109-117.[Abstract/Free Full Text]

This article has been cited by other articles:

				D. L. MARSHALL, N. J. ABRAHAMSON, J. J. AVRITT, P. M. HALL, J. S. MEDEIROS, J. REYNOLDS, M. G. M. SHANER, H. L. SIMPSON, A. N. TRAFTON, A. P. TYLER, et al. Differences in Plastic Responses to Defoliation due to Variation in the Timing of Treatments for Two Species of Sesbania (Fabaceae) Ann. Bot., May 1, 2005; 95(6): 1049 - 1058. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (10) Citing Articles via Google Scholar Google Scholar Articles by Desclaux, D. Articles by Roumet, P. Search for Related Content PubMed Articles by Desclaux, D. Articles by Roumet, P. Agricola Articles by Desclaux, D. Articles by Roumet, P.