Identification of Soybean Genotypes with [N.sub.2] Fixation Tolerance to Water Deficits.
Soil drying sufficient to result in a decline in [N.sub.2] fixation activity is likely to occur during most growing seasons, especially during seed development when the transpiration rates of the crop are high. Observed declines in [N.sub.2] fixation of soybean during early seed fill are probably often linked to developing soil water deficits. Experiments in which soil moisture was carefully maintained at a high level throughout seed development showed no loss of [N.sub.2] fixation activity until near crop maturity (Nelson et al., 1984; Denison and Sinclair, 1985).
An option to increase soybean yield is to eliminate, or at least minimize, the sensitivity of [N.sub.2] fixation to soil drying. An approach for making this improvement would be to identify soybean genotypes that have greater [N.sub.2] fixation tolerance to soil drying, and then to incorporate this trait into commercial cultivars. The fact that there is a large range of diversity among grain legume species in the sensitivity of [N.sub.2] fixation to soil drying (Sinclair and Serraj, 1995) offers support for such an approach. In the comparison of legume species, Sinclair and Serraj (1995) found that those species transporting N to the shoot in the form of ureides, such as soybean, had [N.sub.2] fixation that was sensitive to soil drying while those transporting amides were quite tolerant.
There are indications that genotypic variations in the sensitivity of [N.sub.2] fixation to soil drying might also exist among soybean cultivars. The cultivar Jackson was identified from an initial screen of 28 cultivars as having [N.sub.2] fixation that was less sensitive to soil water deficit (Sail and Sinclair, 1991). Instead of an early decline in [N.sub.2] fixation activity with soil drying, the decline in Jackson was delayed so that it paralleled the decrease in transpiration rate. Serraj and Sinclair (1997) confirmed the superiority of Jackson in a comparison of 18 cultivars. In their study, Jackson also was found to have the lowest level of xylem-sap ureide concentration among eight cultivars that were compared. Interestingly, there were indications of a genetic basis of the tolerance to soil drying because one parent of Jackson, `Volstate', was also found to have [N.sub.2] fixation with some tolerance to soil drying (Serraj and Sinclair, 1996) and to have a low xylem-sap ureide concentration (Serraj and Sinclair, 1997). The other parent of Jackson, `Palmetto', had [N.sub.2] fixation sensitive to water deficit (Serraj and Sinclair, 1996) and relatively high xylem-sap ureide concentration (Serraj and Sinclair, 1997).
To select for [N.sub.2] fixation activity under drought conditions, de Silva et al. (1996) proposed an index based on ureide levels per unit petiole dry weight. Purcell et al. (1998) demonstrated that harvesting petioles from the uppermost fully expanded leaf under field conditions resulted in fairly stable ureide levels both within and across days during the growing season. In field tests involving 24 cultivars, Serraj and Sinclair (1997) found a negative correlation (r = -0.47) between ureide levels in petioles harvested from well-watered plants and the relative amount of [N.sub.2] fixation under dry conditions.
There has not been a wide-ranging search among the soybean germplasm for genotypes with [N.sub.2] fixation tolerance to soil drying. The objective of this investigation was to screen a large number of plant introduction lines in an effort to identify soybean genotypes that exhibit specifically a trait of [N.sub.2] fixation tolerance to soil drying.
MATERIALS AND METHODS
Since direct tests for [N.sub.2] fixation activity with drying soil are time consuming and allow only a few plants to be tested at one time, the approach used in this study involved a three-stage screening procedure. Consequently, an initial screen of a large number of genotypes was narrowed in stages to only a few genotypes for intensive study.
Stage 1: Petiole Ureide Screen
The first-stage screen for [N.sub.2] fixation tolerance to soil-water deficit was the measurement of petiole ureide levels under well-watered conditions. High ureide levels are associated with the sensitivity of [N.sub.2] fixation to soil drying. It is unknown whether soybean germplasm exist that does not transport ureides like those of grain legume species with [N.sub.2] fixation that is highly tolerant of soil water deficits (Sinclair and Serraj, 1995). The entries sampled were accessions in maturity groups V through VIII from the USDA Soybean Germplasm Collection and were grown at Stoneville, MS, on a Bosket fine-sandy loam soil (fine-loamy, mixed thermic, Mollic Hapludalf). Entries were grown in either four-row evaluation plots (Coble et al., 1991), four-row seed increase plots, or one-row comparison plots. Approximately 1000 plant introductions, which were blocked by maturity group, were screened each year. These introductions represented nearly 40 different countries but most of the accessions came from China (40%), Japan (21%), Indonesia (12%), S. Korea (5%), India (5%), and USA (5%). Plots were maintained under well-watered conditions and irrigated as necessary.
In each of three years (1995, 1996, and 1997), petioles were harvested on one day for all cultivars. The harvest day was selected to be at approximately mid-flowering for many of the genotypes and ranged from R1 to R4 (Fehr and Caviness, 1977). Petioles were harvested from the uppermost, fully-expanded leaves from three plants, placed in coin envelopes, and stored in ice chests for approximately 6 h. Subsequently, the petioles were oven dried at 80 [degrees] C and finely chopped. Approximately 30 mg of petiole tissue was extracted in 1 mL of 0.2 M NaOH for 30 min at 100 [degrees] C (de Silva et al., 1996). Ureide concentration was determined with an autoanalyzer based on the method of Van Berkum and Sloger (1983) with the modification that the alkaline hydrolysis step was omitted.
In some cases, petiole ureide concentration was below the calibration range of the autoanalytical system. For these lines subsequent greenhouse experiments were conducted to determine if they were nonureide producing, [N.sub.2] fixing soybeans. This was a potentially important question to resolve given that amide-producing legumes are able to continue fixing [N.sub.2] at soil-water deficits inhibitory to ureide producing legumes (Sinclair and Serraj, 1995). Greenhouse-grown plants were treated with 1 mM allopurinol in 100 mL of N-free nutrient solution (de Silva et al., 1996) once a week following sowing for 4 wk. Allopurinol is an inhibitor of xanthine oxidase (E.C. 220.127.116.11) and blocks [N.sub.2] fixation in ureide-producing legumes but has no effect on amide-producing legumes (Atkins et al., 1988). Plants were observed for sustained growth following the allopurinol treatment.
The genotypes in each year were ranked based on petiole ureide level and about 10% of the lines, which had the lowest petiole ureide levels, were retained for subsequent screening. Also included as checks among the lines for advancement from the 1995 and 1997 field screens were seven and 10 lines, respectively, which fell in the 95th percentile or greater for high petiole ureide concentration.
Stage 2: N Accumulation Screen
Stage 2 involved a direct measure of N accumulation under field conditions of plants subjected to water deficit. Accessions selected in Stage 1 at Stoneville, MS, were tested in 1996, 1997, and 1998 at Gainesville, FL, on Arredondo fine sand soil (loamy, siliceous, hyperthermic Grossarenic Paleudults). This sandy soil had approximately 35 mm of extractable soil water (Sinclair et al., 1987) so rapid imposition of water deficit was possible. The soil was also low in organic content so that virtually all nitrogen accumulated by the soybean plants was a result of [N.sub.2] fixation (Serraj and Sinclair, 1997).
Because of limited seed availability each plant introduction was hand sown in four-row plots with two replicate, randomized blocks. Each plot was 5.5 m in length with 25 cm between rows and 10 cm between seeds with the row. The plots were sown on 5 April 1996, 3 April 1997, and 1 and 2 April 1998. A few genotypes had poor seedling emergence and were discarded. The plots retained in the screen generally had plant populations of 30 to 35 plants [m.sup.-2].
Once the plants had achieved at least 80% of closed canopy, usually by the second week in May, the water-deficit treatment was imposed. Since May is usually quite dry in Gainesville, FL, water-deficit conditions were readily established on this sandy soil within about 4 d after withholding irrigation. Nitrogen accumulation under water-deficit conditions was measured over approximately a 3-wk period during late vegetative growth in a procedure described by Sinclair et al. (1987). The initial harvest was the above ground mass from a 2-m length of the center two rows (1 [m.sup.2]) from each genotype. This first harvest occurred on 10 May 1996, 15 May 1997, and 8 May 1998. All the harvested plant material was dried in an oven at 60 [degrees] C, weighed, and ground. The N concentrations of sub-samples were determined with Leco N determinator (Model F-228, Leco Corp., St. Joseph, MO) by the Soil Testing and Plant Analysis Laboratory at the University of Arkansas.
Since there was little or no rainfall during the experimental period, severe drought stress on plants was avoided by periodic irrigation. Irrigation was done each morning following an afternoon when initial stages of leaf wilting were observed across the treatment block. An intensive array of overhead sprinklers was used to apply uniformly only 10 mm of water on most occasions. The low amount of water application and the high evaporation conditions meant that irrigation was required every second or third day. Consequently, the plants were sustained under mild water-deficit conditions throughout the treatment period.
The experiments were ended each year when a heavy rainfall saturated the soil. The second harvest of the genotypes was done on the day following the saturating rain, which in the 3 yr was 29 May 1996, 2 June 1997, and 29 May 1998. The procedures for this harvest and sample analysis were the same as those used in the first harvest.
Several approaches were considered in examining the data from these experiments to evaluate the relative sensitivity of [N.sub.2] fixation to soil water deficits. The analysis of Sinclair et al. (1987) showed that a critical comparison was the amount of N accumulated relative to the amount of mass accumulated during the water-deficit treatment. Therefore, for each plant introduction this relative value was calculated as the difference in N amount between the two samples divided by the difference in mass harvested for the two sample periods. The unit of this relative value, however, is actually N concentration (g N per g mass) and there appeared to be a bias in this calculation in favor of those genotypes that initially had high N concentrations. Consequently, the calculated value for the difference in N and mass accumulation was divided for each genotype by the N concentration of the first harvest. This calculation resulted in a dimensionless tolerance ratio, which in most cases had a value of less than one. This tolerance ratio was interpreted to reflect the ability of each genotype to sustain [N.sub.2] fixation rates during the water-deficit treatment so as to maintain the initial N concentration of the plant. A high value in this ratio was taken to indicate a potential tolerance of [N.sub.2] fixation to soil drying.
The tolerance ratio was calculated for each plot and the accessions were ranked within each block. However, sustained N accumulation in this Stage 2 screen could have resulted from attributes independent of [N.sub.2] fixation tolerance to water deficits. For example, those genotypes that conserved soil moisture or had deep rooting on this soil would be able to support continued [N.sub.2] fixation even though there was no inherent tolerance of [N.sub.2] fixation to soil drying. Therefore, a third stage of screening was necessary. Approximately 10% of the genotypes that had high tolerance ratios in both blocks from Stage 2 were selected for evaluation in the third stage of screening.
Stage 3: Acetylene Reduction Assay Screen
The acetylene reduction assay (ARA) offered a method to monitor nondestructively the [N.sub.2] fixation activity of nodules in a fixed volume of soil during a drying cycle. In this screen, individual plants were grown in a greenhouse in 10-cm-diam and 30-cm-high polyvinylchloride pots as described by Sall and Sinclair (1991). In addition to the cultivars selected in Stage 2, the sensitive cultivar Biloxi was included with each test group. All plants were maintained in a well-watered condition until they were approximately 5-wk old. At this point, the test of ARA response to soil drying was initiated by fully watering all pots. After draining overnight, the pots were sealed so that acetylene could be flowed briefly through each pot for measurement of ARA.
In this screen, there were generally 10 pots of each genotype. Five plants were maintained as well-watered controls and five plants were subjected to soil drying. Each afternoon over approximately 2 wk, ARA was measured for each pot and the pots were weighed to determine soil water loss. After weighing, the well-watered pots were watered so as to return the pot to a weight of 250 g less than the fully watered weight. The drying pots were watered so as to maintain the net daily decrease in soil water content loss of the pot at about 70 g. As a result, the soil dried slowly in the drying pots and ARA of each plant was measured throughout the drying cycle (Sail and Sinclair, 199t). At the end of the drying cycle, the well-watered plants were harvested, dried, and shoot ureide concentration was measured.
The ARA and water loss of each drying pot on each day was normalized against the average ARA and water loss of the well-watered pots. The normalization allowed variations in ARA and transpiration values resulting from differences in natural environment variations to be eliminated in the analysis. A second normalization was performed by dividing normalized values obtained each day by the average of normalized values for the first 6 d of the drying cycle. The purpose of this second normalization was to buffer plant-to-plant variations. The ARA and transpiration results were plotted as a function of soil water content, which was expressed as FTSW (Sall and Sinclair, 1991). The threshold at which ARA began to decline with decreasing FTSW was calculated by means of the plateau regression procedures of SAS (SAS Institute, Inc., 1988). By definition, the plateau at high FTSW was set equal to one. Linear regression of ARA against FTSW at low FTSW yielded the FTSW threshold for the ARA decline and a 95% confidence interval. The same plateau regression procedures were used to calculate threshold values for the decline in transpiration.
The basis of this screen was the threshold FTSW at which [N.sub.2] fixation activity initially decreased. Those genotypes that showed a decrease in ARA only at low FTSW were ultimately identified as having superior tolerance of [N.sub.2] fixation to soil drying. Commonly, those plant introduction lines that were identified as being superior in [N.sub.2] fixation tolerance to soil drying were retested to confirm the original results.
Stage 1: Petiole Ureide Screen
In the 3 yr of screening, over 3000 plant introductions were measured for petiole ureide content (Table 1). There was a large distribution among the plant introductions in this trait. For example, in the 1995 screen (Fig. 1) the values of petiole ureide content range from 2 to over 100 [micro]mol [g.sup.-1] dry weight. There was a skewed distribution among the plant introductions in this trait with the greatest number of lines being in the range of 5 to 8 [micro]mol [g.sup.-1].
[Figure 1 ILLUSTRATION OMITTED]
Table 1. The number of soybean genotypes screened at each stage and the number of plant introductions identified with drought-tolerant nitrogen fixation.
Year of Stage 1 screen Screen stage 1995 1996 1997 1. Petiole ureide 945 712 1424 2. Nitrogen accumulation 80 28 99 3. Acetylene reduction assay 9 4 5 Tolerant lines 5 1 2
Although the exact range of ureide concentration differed from year to year, the general skewed shape of the distribution was consistent across years. The frequency distribution was normalized by the natural logarithm of ureide concentration (Fig. 1b). Because of the skewed distribution for ureide concentration, the standard deviation of the mean was of the same magnitude as the mean (Table 2). The natural log of ureide concentration of check cultivars for each of the years generally fell between the 5th and 95th percentiles.
The possibility that lines with undetectable petiole ureide levels did not depend on ureide synthesis and transport were tested with the allopurinol treatment. For all lines, allopurinol treatments resulted in chlorosis and decreased plant mass compared with plants without allopurinol. Therefore, across a wide range of improved and unimproved soybean germplasm, ureides appear to be the predominant product of [N.sub.2] fixation in soybean.
Genotypes selected for advancement to the next stage of screening were those lines that had petiole ureide concentration less than 2.6 [micro]mol [g.sup.-1] dw for 1995, 1.36 [micro]mol [g.sup.-1] dw for 1996, and 4.96 [micro]mol [g.sup.-1] dw for 1997. These cut-off values were in the 10th percentile or less (Table 2). Some accessions, which had high petiole ureide concentrations, were also selected as checks for the Stage 2 screen.
Stage 2: N Accumulation Screen
Over 200 plant introductions were subjected to the Stage 2 screen in the 3 yr of trials (Table 1). The results of the N accumulation screen also resulted in a wide distribution among the selected genotypes in each year (Fig. 2; Table 3). In 1997 and 1998 there was a normal distribution of the tolerance ratios, but this was not the case in 1996 apparently because of a skewing towards large tolerance ratios (Fig. 1). Also, there was an overall decrease in the tolerance ratios in 1998 as compared with 1996 and 1997. There is no ready explanation for these differences among years.
[Figure 2 ILLUSTRATION OMITTED]
The intent of this screen was to select those genotypes that sustained [N.sub.2] fixation under field water deficits for further screening in Stage 3. In each test group, approximately 10% of the lines could be readily identified as having clearly high tolerance ratios. These plant introductions tended also to be ones with vigorous plants and good overall growth in the field.
Stage 3: Acetylene Reduction Assay Screen
A total of 18 plant introductions were subjected to the intensive and direct ARA measurement of [N.sub.2] fixation activity in response to soil drying (Table 1). Eight of these selected plant introductions were found to have [N.sub.2] fixation activity that did not decline until fairly severe soil water deficits had developed (Table 4). The FTSW values for the threshold in decline of ARA ranged from 0.11 to 0.28 among these eight genotypes. For seven of the genotypes the threshold was 0.22 or less, and PI507039 exhibited an extremely high level of drought tolerance in [N.sub.2] fixation with a threshold of only 0.11 FTSW. The threshold for ARA for the sensitive, control cultivar Biloxi was 0.33, 0.27, and 0.29 in each of the three screening cycles.
Since soybean generally has a threshold for [N.sub.2] fixation decline with soil drying that is higher than transpiration, an important comparison was the ARA threshold relative to the transpiration threshold. The response to soil drying among these eight genotypes in ARA decline relative to transpiration is illustrated by the results from PI578315B (Fig. 3). In this genotype, the threshold for transpiration was 0.40 FTSW while the threshold for ARA was shifted to a low value of 0.19 FTSW, or a [N.sub.2] fixation threshold that was 0.21 FTSW less than transpiration. Over the eight selected genotypes, the [N.sub.2] fixation threshold ranged from 0.20 to 0.33 FTSW less than the transpiration threshold (Table 4).
[Figure 3 ILLUSTRATION OMITTED]
The shoot ureide concentrations measured for seven of the genotypes were low, including the two plant introductions that had been originally identified in the Stage 1 screen as having extremely high petiole levels (Table 4). The shoot ureide concentrations measured in the Stage 3 screen, however, ranged from 4.3 to 7.5 [micro]mol [g.sup.-1]. These shoot ureide concentrations were substantially less than those of Biloxi which was 15.5 [micro]mol [g.sup.-1] in the first test.
Many studies have identified soybean as having [N.sub.2] fixation that is especially sensitive to soil drying. Indeed, the loss in [N.sub.2] fixation activity when soil is subjected to only modest soil drying is likely to be an important limitation on increasing yielding capability in commercial soybeans. This study has identified eight plant introduction lines that have substantial tolerance in [N.sub.2] fixation to soil drying.
A unique three-stage screen was developed to identify these superior plant introduction lines. The three-stage scheme was needed because direct measurements of [N.sub.2] fixation activity (Stage 3 in the screening process) is laborious and only a limited number of genotypes can be studied. Hence, two preliminary stages of screening were developed to focus the direct measurement of [N.sub.2] fixation response to soil drying on only a few lines with a high probability of success. While the first two stages of the screen (petiole ureide levels and N accumulation) were based on physiological considerations, they were designed basically to identify strong candidates for the intensive ARA studies.
The first-stage screen based on petiole ureide levels allowed many genotypes to be sampled in the field because petioles could be readily harvested and an auto-analyzer system was available for ureide analysis. The original concept for this screen was the elimination of those lines with high levels of petiole ureide, which has been associated with high sensitivity of [N.sub.2] fixation to soil drying. The value of this Stage 1 screen is in doubt, however, on the basis of the ultimate identification of two plant introductions with drought tolerance to [N.sub.2] fixation that were initially found to have high petiole ureide levels. These two lines with high petiole ureide levels were included in the Stage 2 screen in an attempt only to confirm their sensitivity to water deficits.
The identification of two accessions, which originally had high petiole ureide levels, as having [N.sub.2] fixation tolerant to soil drying indicates a confounding problem in a screen based on petiole ureide concentration. Purcell et al. (1998) found that ureides can apparently be stored in petioles, particularly the apical portion. It is possible that a high level of petiole ureide may reflect in some genotypes an ability to sequester and store ureides in the petioles so that they are not involved in the feedback inhibition on nodule activity under water deficit conditions (Serraj et al., 1999b). Subsequent measurements of shoot ureide levels in the greenhouse showed these two accessions had low ureide concentrations (Table 4).
It was hoped that in the Stage 1 screen that soybean germplasm might be identified which contained no petiole ureide because those legume species with high tolerance of [N.sub.2] fixation to soil drying appear not to transport ureides from the nodules (Sinclair and Serraj, 1995). In this sampling of over 3000 plant introductions of soybean, no line was identified that did not produce ureides. Those lines below the calibration range of our analytical system were subsequently shown to depend upon ureide formation in the allopurinol test. Therefore, it appears that ureide synthesis and transport are essential features of [N.sub.2] fixation in soybean.
The second screen of N accumulation under field water-deficit conditions gave a more direct measure of [N.sub.2] fixation tolerance to water deficit. This screen gave a wide distribution in the tolerance ratio among those lines tested (Fig. 2). Those lines that had a high ratio were selected for the intensive ARA experiment. Since about half of the selected lines from Stage 2 showed [N.sub.2] fixation tolerance to water deficit in Stage 3, Stage 2 yielded a high fraction of lines with [N.sub.2] fixation tolerant of water deficits. Those lines that failed to demonstrate a direct tolerance in [N.sub.2] fixation to water deficit in the Stage 3 ARA screen may still be of considerable interest. For example, they may have desirable traits such as deep rooting or water conservation that allowed good performance of [N.sub.2] fixation under the field screen when subjected to drought conditions.
The Stage 3, ARA screen confirmed that, in fact, eight plant introduction lines with superior tolerance of [N.sub.2] fixation to drying soil had been identified. Consequently, the multistage screen was successful in establishing about 0.3% of the original plant introduction population as having superior [N.sub.2] fixation tolerance to soil drying. The eight introductions selected included all of the maturity groups tested (V through VIII) and both determinate and indeterminate types. Four of these introductions came from Japan although Japanese accessions were only 21% of the accessions screened. Our records indicate that the other selected lines came from the tertiary sources of Argentina. Nigeria, India, and Nepal so the origin of the genes involved is uncertain. Although 40% of the accessions tested were from China none were included among the most tolerant.
It is anticipated that these eight genotypes can be used as valuable resources both ill breeding programs and in physiological studies. These superior genotypes can be used as parental germplasm for crosses with cultivars that have good commercial characteristics. Because of the difficulty in directly confirming [N.sub.2] fixation superiority with soil drying on a large number of plants, i.e., Stage 3 screening, breeding efforts may initially have to focus on evaluating overall performance of the progeny under field water-deficit conditions similar to the Stage 2 screen. Once progeny lines have been developed that have demonstrated increased yielding capability under water-deficit conditions, then confirmation of the [N.sub.2] fixation tolerance trait could be attempted.
These eight genotypes could prove to be especially beneficial in physiological investigations of the regulation of [N.sub.2] fixation rates. Since [N.sub.2] fixation is a complex process involving nodule formation, nodule activity, transport between nodule and leaves, and processes in the leaves, it is possible that there are several different mechanisms operating within these genotypes that result in altered [N.sub.2] fixation behavior. An obvious initial question is to resolve whether these genotypes are expressing differing mechanisms that result in superiority of [N.sub.2] fixation tolerance to soil drying. Once the basis for the superiority of these genotypes is resolved, then it may be possible to develop screens directed to a specific trait. Clearly, a more specific screen could be of benefit in a breeding program.
Atkins, C.A., P.J. Sanford, P.J. Storer, and J.S. Pate. 1988. Inhibition of nodule functioning in cowpea by a xanthine oxidoreductase inhibitor, allopurinol. Plant Physiol. 88:1229-1234.
Coble, C.J., G.L. Sprau, R.L. Nelson, J.H. Orf, D.I. Thomas, and J.F. Cavins. 1991. Evaluation of the USDA Soybean Germplasm collection: Maturity groups 000 to IV (PI 490.765 to PI 507.573). USDA Tech. Bull. 1802.
Denison, R.F., and T.R. Sinclair. 1985. Diurnal and seasonal variation in dinitrogen fixation (acetylene reduction) rates by field-grown soybean. Agron. J. 77:679-684.
de Silva, M., L.C. Purcell, and C.A. King. 1996. Soybean petiole ureide response to water deficits and decreased transpiration. Crop Sci. 36:611-616.
Fehr, W.R., and C.E. Caviness. 1977. Stages of soybean development. Special Report No. 80. Iowa State Univ. Coop. Ext. Serv., Ames, IA.
Nelson, D.R., R.J. Bellville, and C.A. Porter. 1984. Role of nitrogen assimilation in seed development in soybean. Plant Physiol. 74: 128-133.
Purcell, L.C., and C.A. King. 1996. Drought and nitrogen source effects on nitrogen nutrition, seed growth, and yield in soybean. J. Plant Nutr. 19:969-993.
Purcell, L.C., R. Serraj, M. de Silva, T.R. Sinclair, and S. Bona. 1998. Ureide concentration of field-grown soybean in response to drought and the relationship to nitrogen fixation. J. Plant Nutr. 21:949-966.
Sall, K., and T.R. Sinclair. 1991. Soybean genotypic differences in sensitivity of symbiotic nitrogen fixation to soil dehydration. Plant Soil. 133:31-37.
SAS Institute, Inc. 1988. SAS/STAT user's guide, Release 6.03. Cary, NC.
Serraj, R., and T.R. Sinclair. 1996. Processes contributing to [N.sub.2]-fixation insensitivity to drought in the soybean cultivar Jackson. Crop Sci. 36:961-968.
Serraj, R., and T.R. Sinclair. 1997. Variation among soybean cultivars in dinitrogen fixation response to drought. Agron. J. 89:963-969.
Serraj, R., T.R. Sinclair, and L.C. Purcell. 1999a. Symbiotic [N.sub.2] fixation response to drought. J. Exp. Bot. 50:143-155.
Serraj, R., V. Vadez, R.F. Denison, and T.R. Sinclair. 1999b. Involvement of ureides in nitrogen fixation inhibition in soybean. Plant Physiol. 119:289-296.
Sinclair, T.R. 1986. Water and nitrogen limitations in soybean grain production. I. Model development. Field Crops Res. 15:125-141.
Sinclair, T.R., R.C. Muchow, J.M. Bennett, and L.C. Hammond. 1987. Relative sensitivity of nitrogen and biomass accumulation to drought in field-grown soybean. Agron. J. 79:986-991.
Sinclair, T.R., and R. Serraj. 1995. Legume nitrogen fixation and drought. Nature 378:344.
Van Berkum, P., and C. Sloger. 1983. Autoanalytical procedure for the determination of allantoin and allantoic acid in soybean tissue. Plant Physiol. 13:511-513.
Abbreviations: ARA, acetylene reduction assay; FTSW, fraction transpirable soil water.
T. R. Sinclair,(*) L. C. Purcell, V. Vadez, R. Serraj, C. Andy King, and R. Nelson
T.R. Sinclair, USDA-ARS, Agronomy Physiology Laboratory, P.O. Box 110965, Univ. of Florida, Gainesville, FL 32611-0965; L.C. Purcell and C.A. King, Agronomy Dep., Univ. of Arkansas, Fayetteville, AR 72703; V. Vadez, Agronomy Physiology Laboratory, P.O. Box 110965, Univ. of Florida, Gainesville, FL 32611-0965; R. Serraj, Faculte des Sciences, BP S 15 Marrakech, Morocco; R. Nelson, USDA-ARS, Soybean/Maize Germplasm, Pathology and Genetics Research Unit, Dep. of Crop Science, Univ. of Illinois, 1101 W. Peabody Dr., Urbana, IL 61801. Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the USDA and does not imply approval or the exclusion of other products that may also be suitable. Received 3 Jan. 1999. (*) Corresponding author (email@example.com).
Published in Crop Sci. 40:1803-1809 (2000).
|Printer friendly Cite/link Email Feedback|
|Author:||Sinclair, T. R.; Purcell, L. C.; Vadez, V.; Serraj, R.; King, C. Andy; Nelson, R.|
|Date:||Nov 1, 2000|
|Previous Article:||Genetic Base of Japanese Soybean Cultivars Released during 1950 to 1988.|
|Next Article:||Repeatability of Large-Scale Germplasm Evaluation Results in Durum Wheat.|