Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 8 September 2006
Published in Crop Sci 46:2104-2110 (2006)
© 2006 Crop Science Society of America
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 Similar articles in ISI Web of Science
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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Longenberger, P. S.
Right arrow Articles by McMichael, B. L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Longenberger, P. S.
Right arrow Articles by McMichael, B. L.
Agricola
Right arrow Articles by Longenberger, P. S.
Right arrow Articles by McMichael, B. L.
Related Collections
Right arrow Crop Physiology & Metabolism
Right arrow Cotton
Right arrow Crop Genetics

CROP BREEDING & GENETICS

Development of a Screening Method for Drought Tolerance in Cotton Seedlings

P. S. Longenbergera,*, C. W. Smitha, P. S. Thaxtonb and B. L. McMichaelc

a Dep. of Soil and Crop Sciences, Texas A&M Univ., College Station, TX 77843-2474
b Delta Research and Extension Center, Mississippi State Univ., Stoneville, MS 38776
c USDA-ARS, SPA, Plant Stress and Water Conserv. Lab., Lubbock, TX 79415

* Corresponding author (PSL6682{at}tamu.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The key to an efficient screening method is the ability to screen large amounts of plant material in the shortest time possible. Due to the complexity of drought tolerance, a quick and effective screen for this trait has yet to be established. The research reported herein was designed to evaluate a screening method for drought tolerance in cotton (Gossypium hirsutum L.) seedlings. Twenty-one converted race stocks (CRS) and two cultivars were evaluated for seedling drought tolerance (SDT) on an individual plant basis. Genotypes were evaluated October–November 2004 and February–March 2005 under greenhouse conditions. Seedlings were subjected to three sequential cycles of drought at 15 d after planting (DAP). Drought cycles consisted of withholding water until the moisture content of indicator ‘Deltapine 491’ (DP 491) plants had an average volumetric water content of 0.07. Plants then were watered to saturation and allowed to drain to field capacity and percent survival recorded after 48 h. Genotypes differed in their percent survival following three consecutive drought cycles. Drought cycles 2 and 3 did not contribute to the separation of genotypes. DP 491 was the most tolerant genotype evaluated. The drought tolerance of the CRS was similar to that of ‘Acala 1517–99’. CRS M-9044–0165 was the most stable genotype, according to an analysis of the difference in percentage of survival for each genotype across the two experiments.

Abbreviations: CRS, converted race stocks • DAP, days after planting • DP 491, ‘Deltapine 491’ • SDT, seedling drought tolerance


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
EFFECTIVE screening methods must evaluate plant performance at critical developmental stages, be completed rapidly, use small amounts of plant material, and screen large numbers of plants (Johnson, 1980). Such screening methods must be incorporated into plant breeding programs to facilitate meaningful genetic improvement. A screening method for drought tolerance in crop plants that fulfills all of these important requirements has eluded researchers to date.

Water is the most important factor limiting crop productivity. The demand for drought tolerant genotypes will be exacerbated as water resources and the funds to access them become more limited. The burden falls on crop physiologists to understand detrimental effects of drought on plant processes and to convey their findings to plant breeders for the development of tolerant genotypes. Difficulties in the past have included the identification of physiological characteristics that are correlated with drought stress that could be used as indicators of drought tolerance. Physiologists are able to measure various plant characteristics that correlate with drought tolerance, such as water use efficiency (Quisenberry et al., 1981; Wright and Dobrenz, 1970; Ray et al., 1974), root characteristics (Pace et al., 1999; Basal et al., 2003; Ball et al., 1994; Cook and El-Zik, 1992), detached leaf water loss (Basal et al., 2005; Roark et al., 1975), leaf water potential (Quisenberry et al., 1985; Kaul, 1969), stomatal characteristics (Quisenberry et al., 1982; McDaniel, 2000), and osmotic adjustment (Nepomuceno et al., 1998; Oosterhuis and Wullschleger, 1987), but these tests are either too tedious or too time consuming for plant breeders to evaluate large segregating populations.

Dryland-produced crops commonly are subjected to periodic water deficits during any particular growing season. Since society is reliant on these plants for food and fiber, their ability to survive and produce during periods of water deficit is of interest to scientists as well as society in general. Measuring drought tolerance of individual plants or established genotypes under field conditions is difficult due to variation in weather conditions and between and within field soil type variation. This has led to efforts of drought simulation studies in greenhouses and growth chambers.

A growth chamber programmed to simulate the first 10 d in August in the arid Southwest was used to select for seedling drought tolerance among three species from two genera of range grasses (Wright, 1964). Wright's growth chamber selection technique separated the species into the same order as their natural range performance. The ability to select among large numbers of seedlings and the capacity to control environmental conditions were cited by the author as the most important benefits of his protocol. Wright and Jordan (1970) used Wright's 1964 method to select among boer lovegrass (Eragrostis curvula Nees) seedlings for drought tolerance. The most tolerant selection of 16 clones evaluated was assessed for range performance and was found to be superior to the check cultivar.

Wright and Brauen (1971) attempted to identify associations between seedling drought tolerance, as determined by the programmed chamber method of Wright (1964), and plant characteristics. Thirty-six lovegrass lines and a commercial cultivar, Lehmann, were evaluated using the chamber method, and surviving plants were evaluated for plant characteristics such as growth habit, foliage color, anther color, chromosome number, seed weight, and seed dormancy. None of the characteristics studied provided adequate association with seedling drought tolerance that would validate its use as a selection criterion.

The drought tolerance of forage grasses was appraised in a growth chamber by watering to field capacity and then withholding water until the soil-filled trays containing the plants reached a predetermined weight (Tischler et al., 1991). Although this method gave consistent results, the authors expressed concern that the tray system did not allow for the extraction of water by long, deep roots in a deep soil profile and that additional techniques were needed to assess such drought-avoidance mechanisms.

Lichthardt and Weaver (1985) assessed range grass seedlings with varying leaf numbers (varying age) for drought tolerance by monitoring their survival after a 2-wk-long drought. The authors found significant differences among the four genotypes tested when plants with more than four leaves were tested. The authors stressed that measuring drought tolerance in pots does not test all aspects of drought tolerance since plants have limited soil to explore for water.

Eight corn lines were evaluated under controlled heat conditions in an environmental chamber (Hunter et al., 1936). Injury ratings and subsequent rankings of the lines were similar for both the controlled heat experiments and previous field observations. The authors stated that the development of a simple seedling test to determine drought tolerance would be a valuable tool for corn breeders who consider tolerance to hot and dry conditions a top priority. Heyne and Brunson (1940) evaluated several corn strains with known field-based drought classifications and their F1 progeny from crosses for their response to controlled high temperature treatments to confirm the field-based drought tolerance ratings found during field studies as indicated by plant injury, recovery, and survival. Essentially the same order of relative resistance among the strains was found in the controlled heat experiments as in mature plants subjected to drought and heat in the field. The authors summarize that heat tolerance was heritable and that in most cases it was controlled by dominant gene action.

Sammons et al. (1978) screened 20 soybean cultivars believed to vary in drought tolerance. The experiment was conducted in a growth chamber and pots were premoistened to three different soil water potentials, –0.07 MPa (control), –0.3 MPa (intermediate moisture stress), and –0.89 MPa (severe moisture stress). Water potential, leaf area and dry weight, and photosynthesis were measured. Significant cultivar x moisture treatment interactions were found for all plant characteristics, and cultivars were not consistently categorized by the variables measured. The same authors later screened the same 20 soybean cultivars in large wooden boxes (1.5 by 3.7 by 0.6 m) under greenhouse conditions (Sammons et al., 1979). Water was withheld for the duration of the experiment (30 d). Leaf lamina expansion rate and plant growth rate were measured, neither of which consistently classified the cultivars relative to drought tolerance. The authors concluded that a combination of characteristics must be evaluated to assess drought tolerance.

Researchers often utilize multiple plant boxes to establish plants for drought evaluation. Boxes (1.5 by 3.7 by 0.6 m) were used to differentiate between seven alfalfa cultivars and three germplasm sources for forage yield and root characteristics under three moisture treatments under shadehouse conditions (Salter et al., 1984). Forage yield decreased by 30 and 50% in the intermediate and low moisture treatments, respectively, compared to the high moisture treatment. There were no differences among genotypes in the low moisture treatment. Root fibrousness increased and root weight decreased with increased moisture stress. All root characteristics correlated with forage yield and the authors determined that the added time and expense of root excavation was not worth the additional information.

Fifteen-day-old cotton seedlings were subjected to four, 4-d drought cycles in a growth chamber to determine drought tolerance (Penna et al., 1998). The authors used soil-filled trays (50 by 35 by 9 cm) containing 12 rows of plants with 12 plants per row. Significant differences in seedling survival were found but results were not consistent across repeated experiments. An unfortunate aspect of using boxes or trays for drought evaluation includes inconsistent plant spacing, varying rates of germination, and variability in root characteristics. Evaluating individual plants or planting single genotypes per container will avoid some of these issues.

The objectives of this study were to (i) develop a protocol to evaluate cotton seedlings for drought tolerance on an individual plant basis and (ii) compare a number of CRS of upland cotton with two cultivars representing diverse germplasm pools for seedling drought tolerance.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Twenty day-neutral converted cotton accessions (McCarty and Jenkins, 1993) were tested for seedling drought tolerance (Table 1). The photoperiodic primitive race stocks were crossed with Deltapine 16 and day-neutral plants were selected in the F2 generation. Day-neutral progenies were then backcrossed four times to the primitive parent. Selection for day-neutrality occurred in the F2 following each backcross. Each CRS has a designation beginning with the letter M, followed by four digits which indicate the year of release and the BCnFn generation at the time of release. The last four digits correspond to the Texas (T-) Cotton Germplasm Collection accession number of the unconverted primitive stock (Basal et al., 2003). Seed of all CRS used in this study were produced at Weslaco, TX, in 2003.


View this table:
[in this window]
[in a new window]
  		Table 1. Day neutral accession number, PI number of the original accession, race designation, and origin of converted race stocks screened for seedling drought tolerance.

 
All CRS, with the exception of M-8844–0096, were identified as the 10 most SDT and 10 least SDT in a survey of 79 CRS using an alternate protocol at the Cotton Improvement Laboratory in 2002 (Darren G. Jones, unpublished data, 2002). M-8844–0096 was identified as putatively heat tolerant in a previous field experiment testing seed set efficiency during the extremely hot growing season of 1998 in College Station, TX (unpublished data).

Two cultivars, Acala 1517–99 and DP 491, were included to represent two distinct germplasm pools and not based on their drought tolerance or susceptibility. Acala 1517–99 was released by the New Mexico Agricultural Experiment Station in 1999 (Cantrell et al., 2000). It was developed to produce high yield and superior fiber quality in New Mexico under conventional irrigated culture. The origin of Acala 1517–99 was from a single plant selection from experimental B2541 by C.L. Roberts. The cross B742/E1141 gave rise to B2541. B742 was derived from Acala 9136/250, with Acala 9136 reportedly having significant introgression from G. barbadense L. cv. Tanguis. E1141 and 250 are of unknown origin. DP 491 was released by the Delta and Pine Land Company (Scott, MS) in 2002. It is marketed as a picker cultivar with high gin turnout and low micronaire suitable for all states in the Cotton Belt with the exception of Arizona.

A preliminary experiment was performed to determine the LD50 volumetric water content at which 50% of DP 491 plants would not recover from seedling drought stress. Volumetric water content is the volume of water per volume of soil and is expressed as a fraction devoid of units. One hundred and sixty-eight Ray Leach (Canby, OR) cone-tainers (3.8 cm diameter by 14 cm depth) were filled by volume with 83 cm3 of fritted clay (Absorb-N-Dry, Flatonia, TX). Fritted clay was selected as the media for this protocol due to its ability to easily accept water after drying events. Van Bavel et al. (1978) characterized the water relations associated with fritted clay. The experiment was performed in an environmental growth chamber with a 12-h daylength, 32°C day/25°C night and 80% relative humidity at the Norman E. Borlaug Center for Southern Crop Improvement, College Station, TX. Two DP 491 seeds were sown per cone-tainer and thinned to one plant per cone-tainer after germination. cone-tainer trays were rotated daily to minimize variation due to microclimates in the growth chamber. All cone-tainers were watered to saturation and allowed to drain to field capacity daily until 15 DAP (plants had two cotyledons and one true leaf). Twenty-eight of the cone-tainers were weighed periodically to monitor water loss. At approximately 120 h post field capacity a group of 14 cone-tainers was rewatered and percentage of recovery recorded after 48 h. Recovery was defined as having at least one turgid leaf and a live apical meristem. Groups of 14 cone-tainers were rewatered after various time durations post field capacity until all remaining cone-tainers had been rewatered with the last group being rewatered at 245 h post field capacity. The plot of percentage of survival versus volumetric water content generated from this experiment indicated that an average volumetric water content of 0.08 was required for 50% recovery of DP 491 (Fig. 1 ).


Figure 1
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 1. Percentage of survival of Deltapine 491 rewatered at specific volumetric water contents under growth chamber conditions at College Station, TX, in 2004.

 
A preliminary SDT screen was performed under greenhouse conditions following the procedure outlined in the following paragraph. Data (not shown) from this preliminary screen indicated that the LD50 volumetric water content for this study should be adjusted to 0.07.

Two experiments were performed. Experiment 1 was conducted October–November 2004 and Experiment 2 was conducted February–March 2005. Experiments 1 and 2 were performed under greenhouse conditions with approximately 31/24°C (day/night), at the Norman E. Borlaug Center for Southern Crop Improvement, College Station, TX. Cotton seedlings were evaluated for drought tolerance on an individual plant basis. Cone-tainers were filled by volume with 83 cm3 of fritted clay. Two seeds were sown per cone-tainer and thinned to one plant per cone-tainer after germination. The experimental design was a randomized complete block with four replications. Each replication consisted of 14 plants of each converted race stock and Acala 1517–99. Each replication included 56 plants of DP 491 so that a better estimate of average volumetric water content could be obtained.

Trays holding the cone-tainers were rotated daily to minimize variation caused by microclimates in the greenhouse. All cone-tainers were watered to field capacity daily as previously described until 15 DAP when the seedlings were subjected to three sequential cycles of drought. Drought cycles consisted of withholding water until the moisture content of the 56 "indicator" cone-tainers, containing DP 491, had an average volumetric water content of 0.07. The duration of drought cycles 1, 2, and 3 for Experiment 1 was 305.5, 364, and 314 h post field capacity, respectively. For experiment 2 cycles 1, 2, and 3 were 297, 328.5, and 338 h post field capacity in length, respectively. Plants were then watered to field capacity and percentage of survival recorded after 48 h. DP 491 plants that did not survive a drought cycle did not contribute to the volumetric water content estimate in future cycles.

Surviving plants were defined as having at least one turgid leaf and a live apical meristem. At the start of drought cycle 1 each plant had two cotyledons and one true leaf. Some plants shed their cotyledons in response to the drought stress. Plants did not develop more than two true leaves throughout the duration of all drought cycles. Percentage of survival following each drought cycle was determined by taking the number of survivors following the drought divided by the number of plants that were alive at the start of the drought cycle.

Since the response variable was binomial in nature, that is, each plant was alive or dead after each drought cycle, the data were analyzed with mixed-effects logistic regression in SAS software using the GLIMMIX procedure (release 9.1.3) (SAS Institute, 2004). The analysis included experiment, genotype, and drought cycle as fixed effects and replication nested in experiment as a random effect.

Denominator degrees of freedom (DDF) were calculated using the containment method. DDF for fixed effects that are found in the syntax of the random effect, in this case, experiment, are the smaller of the rank contributions to the [X Z] matrix of the random effect, in this case replication nested in experiment. Fixed effects not found in the random effect are assigned DDF equal to n-rank [X Z]. There were a total of 624 observations in this experiment but only 617 contributed to n since seven observations had zero percent survival. Rank is equal to the number of independent columns in a matrix.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
The ANOVA indicated that percentage of survival differed among the 23 genotypes (Table 2). The combined analysis for three drought cycles and two experiments indicated a difference among drought cycles. The average percent survival for drought cycles 1, 2, and 3 across both experiments was 47, 94, and 89, respectively, verifying that the largest amount of plant mortality occurred during the first drought cycle in each experiment. Individualized analysis of drought cycles indicated differences in survival among genotypes following one drought cycle but not drought cycles 2 or 3. No further information was gained from additional drought cycles. The relationship between percentage of survival after drought cycle 1 and mean number of survivors after drought cycle 3 had an R2 of 0.94 which further verified the lack of additional information gained from drought cycles 2 and 3 (Fig. 2 ). Plant breeders would be interested in making selections from the final survivors.


View this table:
[in this window]
[in a new window]
  		Table 2. Variance analysis for percentage of survival of 21 converted race stocks and two cotton cultivars after three cycles of drought in two experiments conducted under greenhouse culture at College Station, TX, in 2004 and 2005.

 

Figure 2
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 2. Mean number of surviving plants of 21 converted race stocks and two cotton cultivars following three consecutive drought cycles versus percentage of survival after one cycle of drought.

 
The ANOVA indicated significant interaction among the genotypes tested and seedling drought tolerance experiment suggesting that genotypes should not be pooled across experiments. That interaction is addressed below. However, the F value for genotype is approximately 5 times that of the interaction so pooling over experiments appears appropriate to provide an overview of the relative SDT of these genotypes.

To avoid inflation of the survival data by including cycles 2 and 3, only the means from drought cycle one were separated at P < 0.05 (Table 3). DP 491 was the most SDT genotype with 82% survival while Acala 1517–99 averaged 50% survival. None of the CRS were more tolerant or susceptible than Acala 1517–99, and all were less tolerant of seedling drought than DP 491. M-8744–0175 was the most tolerant CRS with a percentage of survival of 60 but was not different than 16 of the other CRS and Acala 1517–99.


View this table:
[in this window]
[in a new window]
  		Table 3. Percentage of survival of 21 converted race stocks and two cotton cultivars after one cycle of drought in two experiments conducted under greenhouse culture at College Station, TX, in 2004 and 2005.

 
Since the generalized linear mixed model output does not supply the mean square error and degrees of freedom from the experiment x genotype interaction, a general linear model (GLM) analysis was performed on the logit transformations of the survival data collected after one drought cycle. The logit transformation was selected to normalize the data since it is also the transformation performed in the GLIMMIX procedure. The logit transformation is used in logistic regression since it has many of the desirable properties of a linear regression model. It is linear in its parameters, may be continuous, and may range from –{infty} to +{infty}, depending on the range of the response variable (Hosmer and Lemeshow, 1989).

The GLM indicated, as did the GLIMMIX procedure, that percentage of survival of some genotypes was differential across the two experiments. A uniform response among genotypes in direction or magnitude across experiments was not observed (Table 4). Differences in direction are illustrated in the response of seven of the genotypes having a negative difference in percent survival across experiments, indicating that their percentage of survival was lower in experiment two than it was in experiment one. The remaining 16 genotypes had a higher percentage of survival in experiment two than in experiment one.


View this table:
[in this window]
[in a new window]
  		Table 4. Difference in percentage of survival of 21 converted race stocks and two cotton cultivars after one cycle of drought in two experiments conducted under greenhouse culture at College Station, TX, in 2004 and 2005.

 
CRS M-9044–0165 was the most stable genotype across experiments with a difference in percentage of survival between experiments of only 1%. Only M-9044–0206 differed from M-9044–0165 in stability across these two experiments as indicated by genotype x experiment interactions. DP 491, with a difference of 15, was more stable across the two experiments than Acala 1517–99 which had a difference of –22. Thirteen of the CRS were also more stable than Acala 1517–99, while none of the CRS showed more stability than DP 491. M-9044–0061, M-9044–0007, and M-9044–0017 showed less stability than DP 491.

Visual observations during both experiments indicated that the CRS and Acala 1517–99 grew at a slightly faster rate than DP 491. DP 491 plants seemed to maintain their first true leaf and their cotyledons during the drought cycles while the CRS and Acala 1517–99 continued to elongate and put on two true leaves. This continued development during drought may have led to the lower survival percentages observed among the CRS and Acala 1517–99. Although a higher percentage of DP 491 seedlings survived the seedling drought imposed in this study, it appears they did so at the expense of growth and development. Both types of drought tolerance exhibited by the genotypes can be adventitious depending on the specific situation. For instance, the tolerance exhibited by DP 491 would be beneficial if one severe drought were to be experienced and the plants had time during the remainder of the season for normal development and production. However, if numerous short drought periods were to occur, genotypes like the CRS and Acala 1517–99 could be preferred since they may continue to grow during the drought periods, and thus establish a plant structure commensurate with maximum yield.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
A protocol was developed to screen cotton seedlings for SDT on an individual plant basis. Evaluation of genotypes under growth chamber conditions may allow for better separation since temperature and humidity can be controlled more precisely, thus reducing or eliminating the genotype x environment interaction. Two out of the three drought cycles were longer in duration (hours post field capacity) in Experiment 1 than in Experiment 2, which also may have added to the genotype x experiment interaction. Although the plants were rewatered when the indicator DP 491 cone-tainers reached 0.07 volumetric water content for each drought cycle, the difference in the length of the cycle may have triggered different physiological responses among the genotypes. Further trials of the protocol and additional evaluation of the genotypes by other methods may further validate the credibility of this method as a dependable screening tool.

Drought cycles 2 and 3 did not provide additional information relative to the SDT of the genotypes in this study. Therefore, one drought cycle is sufficient to separate cotton genotypes for SDT, although additional cycles could be desirable in a true breeding study designed to identify individual plants for advancement to future screens.

Due to the limited fritted clay volumes used in this test, genotypes showing drought tolerance in nonlimiting environments due to their ability to access water deep within the soil profile may not have been accurately ranked by this technique. Field evaluations or the use of larger containers may confirm or refute this complication.

DP 491 was consistently ranked as the most drought tolerant genotype tested with this method. It is unclear if this ability to survive seedling drought translates into drought tolerance at other stages of growth.

The SDT protocol was developed as a new method for the selection of drought tolerant individual plants among various genotypes subjected to sequential drought cycles. It followed the lead of the research performed by Penna et al. (1998). There are some important differences between this protocol and previous soil filled tray methods. Individual plants were evaluated in cone-tainers to alleviate the competition between plants when they share a common soil volume. Fritted clay was selected as the media for this protocol due to its ability to easily accept water after drying events. Lastly, the use of indicator plants and their volumetric water content was an attempt to remove the subjective nature of visual wilting symptoms as the measure of drought cycle duration. Visual observations had been used in previous experiments at the Cotton Improvement Laboratory (Darren G. Jones, unpublished data, 2002).

Received for publication March 27, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 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 Similar articles in ISI Web of Science
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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Longenberger, P. S.
Right arrow Articles by McMichael, B. L.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Longenberger, P. S.
Right arrow Articles by McMichael, B. L.
Agricola
Right arrow Articles by Longenberger, P. S.
Right arrow Articles by McMichael, B. L.
Related Collections
Right arrow Crop Physiology & Metabolism
Right arrow Cotton
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal 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