Crop Science
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 20 May 2008
Published in Crop Sci 48:903-910 (2008)
© 2008 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
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 Lorenzana, R. E.
Right arrow Articles by Bernardo, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Lorenzana, R. E.
Right arrow Articles by Bernardo, R.
Agricola
Right arrow Articles by Lorenzana, R. E.
Right arrow Articles by Bernardo, R.
Related Collections
Right arrow Maize
Right arrow Crop Genetics

Genetic Correlation between Corn Performance in Organic and Conventional Production Systems

Robenzon E. Lorenzana and Rex Bernardo*

Dep. of Agronomy and Plant Genetics, Univ. of Minnesota, 411 Borlaug Hall, 1991 Buford Cir., St. Paul, MN 55108

* Corresponding author (bernardo{at}umn.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Cultivars bred under conventional production systems may not be optimum for organic production systems. Our objective was to determine if, on the basis of quantitative genetic parameters, separate corn (Zea mays L.) breeding programs are needed for organic and conventional production systems. Testcrosses of 119 intermated B73 x Mo17 recombinant inbreds were evaluated in organic and conventional systems in both Waseca and Lamberton, MN, in 2006. Differences in trait means between the two production systems were significant for grain moisture, plant height, and ear height but not significant for grain yield, root lodging, stalk lodging, and stay green. The organic system led to a smaller testcross genetic variance for grain yield and higher testcross genetic variances for all other traits. The organic system led to a lower heritability for grain yield and a higher heritability for root lodging, stay green, and ear height. Genetic correlations for performance in the two production systems were 0.84 for grain yield; greater than 0.90 for grain moisture, plant height, and ear height; and about 0.50 for root lodging and stay green. The predicted ratio between the correlated response and direct response to selection in the organic system was near 1.0 for grain yield and moisture and considerably less than 1.0 for other traits. These results suggest that high-yielding cultivars for organic systems can be developed largely by screening conventional inbreds and hybrids for their performance under organic systems.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
MANY ORGANIC FARMERS in the United States are constrained to plant cultivars developed for conventional production systems because of a lack of cultivars bred specifically for organic production (Colley and Dillon, 2004; Carr et al., 2006). Conventional cultivars, bred for high yield in high-input conventional production systems, may not be well-adapted to organic production systems (Lammerts van Bueren et al., 2003; Kirschenmann, 2004; Brummer, 2004; Carr et al., 2006; Lammerts van Bueren and Verhoog, 2006). Breeding programs for organic production would require major investments in time, resources, and labor. The advantages of selection under organic conditions, if any, need to be determined (Tracy, 2004; Lammerts van Bueren and Verhoog, 2006), especially because the organic seed market is still relatively small (Colley and Dillon, 2004) and unprofitable for the major private breeding companies and funds to support public breeding are limited (Carr et al., 2006).

To date, the need for a separate breeding program for organic corn (Zea mays L.) cannot be objectively assessed because the quantitative genetics of corn performance under organic conditions has not been studied. Our objective was to determine if, on the basis of quantitative genetic parameters, separate corn breeding programs are needed for organic and conventional production systems. In this study, we compared the mean, genetic variance, heritability, and predicted response to selection for corn performance in both organic and conventional production systems and estimated the phenotypic and genetic correlations between corn performance in the two production systems. Estimates of genetic variance and heritability would be useful in predicting the response to selection (Dudley and Moll, 1969) in organic and in conventional systems. Estimates of the genetic correlation between performance in two environments (Falconer, 1952; Robertson, 1959; Eisen and Saxton, 1983; Yamada et al., 1988), in this case between performance in organic versus conventional systems, would indicate the extent of genotype by production system interaction. Performance in an organic system can be enhanced by selection under the organic system itself (i.e., direct selection). Performance in an organic system may also be enhanced, at least to some extent, by selection under the conventional system (i.e., correlated or indirect response). From the estimates of heritability and genetic correlation, the predicted ratio of correlated response and direct response can be calculated. Indirect selection may be superior to direct selection if the trait heritability is lower in the target environment (i.e., organic system) than in another environment where indirect selection is to be conducted (i.e., the conventional system), and the genetic correlation between performance in the two environments is high (Falconer, 1952; Falconer and Mackay, 1996).


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Plant Materials and Field Evaluation
In 2005 recombinant inbreds derived from the intermated B73 x Mo17 population (Lee et al., 2002) were crossed to LH 295, an elite Monsanto inbred widely used in the northern Corn Belt. Sufficient seeds for multilocation yield trials were obtained from 119 crosses.

In 2006 yield trials were conducted in conventional and organic production systems at the University of Minnesota Southern Research and Outreach Center in Waseca and at the University of Minnesota Southwest Research and Outreach Center in Lamberton. The conventional and organic sites in each location were within 1 km of each other. The organic site at Waseca was in the process of organic certification whereas the organic site at Lamberton has been certified organic since 1998. The 119 B73 x Mo17 recombinant-inbred testcrosses were randomly divided into four sets, three with 30 testcrosses each and one with 29 testcrosses. Two hybrids were added as checks in all four sets, and an extra check was added to the set with 29 testcrosses, for a total of 32 entries in each set. Each set was evaluated in a randomized complete block design with two replications. All seeds used were untreated.

The entries were planted in two-row plots, 4.57 m long and spaced 0.76 m apart, at a plant population density of 78,000 plants ha–1. In the conventional sites, plots were planted in late April (Table 1 ) in accordance with the usual planting dates for the region. In the organic sites, planting was delayed by about 3 wk to allow mechanical weeding of early flushes of weeds. Standard agronomic management practices for high-input conventional systems were applied in the conventional sites in both locations. Organic management practices for the organic sites at each location differed in their previous crop, fall cultivation, and organic fertilizer applied.


View this table:
[in this window]
[in a new window]

  		Table 1. Selected site characteristics and management practices for the yield trials of B73 x Mo17 recombinant inbred x LH295 testcrosses under conventional and organic production systems at Waseca and Lamberton, MN, in 2006.{dagger}

 
Data obtained from each plot were grain yield (in Mg ha–1, adjusted to 155 g H2O kg–1), grain moisture (in g kg–1), root lodging (percentage of plants leaning 45° or more from the vertical), stalk lodging (percentage of plants with stalks broken below the ear), stay green (rated from 1 [high] to 9 [low]), plant height (distance from the soil surface to the tip of the tassel, in cm), and ear height (distance, in cm, from the soil surface to the node below the top ear). Given the differences in planting date between the production systems (Table 1), stalk lodging, root lodging, and stay green in the organic sites were measured 17 d after they were measured in the conventional sites. For plant height and ear height, data were available from the conventional and organic sites in Waseca only. The plots were also evaluated for leaf diseases but no significant disease incidence was observed at any of the sites.

Trait Means, Variance Components, and Heritabilities
Means of the testcrosses in the conventional system and organic system across locations were calculated for each trait. To estimate variance components for each trait in each production system, data from all checks were excluded and an analysis of variance (ANOVA) was performed using PROC GLM in SAS/STAT (SAS Institute, 2004), with locations and testcrosses considered as random effects. Testcross genetic variance (VTC), testcross by location interaction variance (VTCxLoc) and error variance (VError) were obtained by equating the observed mean squares from the ANOVA to their expectations and solving for the desired variance components. For each trait in each production system, heritability on a testcross mean basis (h2) was calculated (Bernardo, 2002). Approximate 95% confidence intervals (CI) for VTC, VTCxLoc, and h2 were calculated according to Knapp et al. (1987), and an exact confidence interval for VError was calculated according to Oehlert (2000). Variance component and heritability estimates were declared significantly different from zero if the approximate 95% CI did not include zero.

Genetic and Phenotypic Correlations
For the traits with data from both Waseca and Lamberton, the genetic correlation (rA) between the conventional and organic production systems was estimated as described by Eisen and Saxton (1983) from an ANOVA combined across locations and production systems. This combined ANOVA also allowed the significance testing, via F-tests, of the differences in trait means between the conventional and organic production systems. Expected mean squares and observed mean squares were obtained using PROC GLM in SAS/STAT (SAS Institute, 2004). From the variance component estimates, rA estimates were calculated as (Eisen and Saxton, 1983)

Formula 1[1]
where V'TC is the testcross genetic variance across locations and production systems (denoted by the prime symbol to distinguish it from VTC within production systems); V'TCxLoc is the variance due to the testcross genotype by location interaction across production systems; VTCxSys is the variance component due to the testcross genotype by production system interaction; VTCxLocxSys is the variance due to the testcross genotype by location by production system interaction; and K is a correction factor for removing bias due to heterogeneity of genetic variances among location by production system combinations. The K value was estimated as (Eisen and Saxton, 1983)

Formula 2[2]
where i is the number of locations, t within production systems); VTC lp (or VTC lp') is the testcross genetic variance from an ANOVA for the production system p (or p') at location l.

For plant height and ear height, the genetic correlation between the conventional and organic production systems in Waseca was estimated as (Yamada et al., 1988)

Formula 3[3]
where V''TC was the testcross genetic variance across production systems in Waseca (denoted by the double prime symbol to distinguish it from VTC within production systems and V'TC across locations and production systems) and V''TCxSys was the variance component due to the testcross genotype by production system interaction in Waseca. The genetic correlation, in this case, is actually the intraclass correlation coefficient, which gives the lower limit of the actual genetic correlation coefficient (Yamada et al., 1988).

An approximate standard error (SE) for rA was calculated (Falconer and Mackay, 1996). An approximate 95% CI for rA was calculated as rA ± 2 SE(rA) (Lynch and Walsh, 1998; Holland, 2006), and an rA estimate was declared significantly different from zero at P = 0.05 if the approximate 95% CI did not include zero.

Phenotypic correlations (rP) were calculated as simple product-moment correlation coefficients between testcross means in conventional and organic production systems across locations. The SE of rP was obtained (Lynch and Walsh, 1998).

Effectiveness of Indirect Selection
The predicted response to direct selection in the organic system was denoted by ROrg. Improvement in performance in the organic system though selection in the conventional system (i.e., correlated or indirect response) was denoted by CROrg. For each trait, assuming that the selection intensities in both production systems were equal, the efficiency of indirect selection under the conventional system to improve the performance in the organic system was calculated as (Falconer and Mackay, 1996)

Formula 4[4]
where hConv is the square root of heritability in the conventional system, and hOrg is the square root of heritability in the organic system. Indirect selection was expected to be more effective than direct selection if the ratio of the indirect response to the direct response to selection was greater than 1.0 (Falconer, 1952; Falconer and Mackay, 1996).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Trait Means
Mean grain yield across locations in the organic system, although about 5% or 0.6 Mg ha–1 lower, was not significantly different from the mean grain yield in the conventional system (Table 2 ). Grain yields were high (>11 Mg ha–1) in both systems because of good growing conditions in 2006, high fertility levels, and absence of stresses such as weeds, insect pests, and diseases in the organic and conventional systems at the two locations. Differences in means between the organic and conventional production systems across locations for root lodging, stalk lodging, and stay green were likewise not statistically significant.


View this table:
[in this window]
[in a new window]

  		Table 2. Trait means of the B73 x Mo17 recombinant inbred x LH295 testcrosses in conventional and organic production systems across two Minnesota locations in 2006.

 
Delate et al. (2003) also found no significant differences between yields in organic and conventional systems, whereas Porter et al. (2003) reported that corn yields under organic management were 7 to 9% lower compared to yields in conventional high-input systems. Lower corn yields in organic production were reported by Pimentel et al. (2005) during the transition from a conventional to an organic system, but after the transition period, corn yields in the organic system approached those of the conventional system. Corn yields in organic systems were observed to be higher than in conventional systems in drought conditions (Lotter et al., 2003; Pimentel et al., 2005).

The absence of significant differences in mean grain yield between the two production systems across locations may have been due to the favorable growing conditions in both locations and the genetic heterogeneity of the materials used in this study, which offset the expected effects on yield of the delayed planting of the organic trials. Delayed planting has been shown to result in reduced corn grain yield in the northern Corn Belt (Alessi and Power, 1975; Imholte and Carter, 1987; Benson, 1990; Lauer et al., 1991). In southern and central Minnesota, a 3-wk delay in planting after the optimal planting date of late April to early May has been shown to result in 5% (for early season hybrids) to 10% (for late season hybrids) average reduction of corn yields (Hicks et al., 1991).

In contrast, the difference in mean grain moisture between the organic and conventional production systems across locations was significant. In Waseca, the organic system lead to significantly higher plant height and ear height compared to the conventional system. These differences can be attributed, at least in part, to the delayed planting of the organic trials. Delayed planting generally results in increased grain moisture (Alessi and Power, 1975; Imholte and Carter, 1987; Lauer et al., 1991). Later planting has also been shown to result in increased plant height in corn (Imholte and Carter, 1987; Nafziger et al., 1991).

Variance Components and Heritabilities
For grain yield, VTC was 80% larger in the conventional system than in the organic system (Table 3 ). For all other traits, VTC was generally larger in the organic system than in the conventional system, although the CI for the VTC estimates in the two production systems were non-overlapping only for grain moisture and stay green (Table 3). The testcross by location variance, VTCxLoc, was significant only for root lodging and stay green for both systems, and for grain moisture in the organic system. The VError estimate for grain yield was slightly higher in the organic system; comparable between the two systems for plant height, ear height, and stay green; significantly smaller in the organic system for root lodging; and larger in the organic system for grain moisture and stalk lodging.


View this table:
[in this window]
[in a new window]

  		Table 3. Variance component and heritability (testcross mean basis) estimates in B73 x Mo17 recombinant inbred x LH295 testcrosses evaluated under conventional and organic production systems across two Minnesota locations in 2006.

 
Estimates of heritability on a testcross-mean basis for grain yield were numerically higher in the conventional system (h2 = 0.65) than in the organic system (h2 = 0.52), although their CI overlapped (Table 3). The higher h2 for grain yield in the conventional system was due to both a larger VTC and a smaller VError compared to the organic system. For grain moisture, h2 estimates were high (>0.80) for both conventional and organic systems. Estimates of h2 for root lodging and stay green were significantly different from zero only in the organic system, whereas h2 estimates for stalk lodging were nonsignificant in each system. The h2 estimates for ear height were considerably higher in the organic system than in the conventional system, whereas estimates of h2 for plant height were comparable between systems. Overall, the results for genetic variances and heritability estimates indicated that genetic differences for grain yield may be better expressed under a conventional system, whereas genetic differences for root lodging, stay green, plant height, and ear height may be better expressed under an organic system.

Phenotypic Correlations, Genetic Correlations, and Efficiency of Indirect Selection
For all traits except stalk lodging, estimates of phenotypic correlation (rP) on a testcross mean basis between the organic and conventional production systems were significantly different from zero (Table 4 ). The rP for grain yield was moderate (rP = 0.39). Cultivar trials, wherein the entries are typically derived from different populations, allow the estimation of correlation between mean cultivar performance in organic and conventional production systems but do not allow the estimation of the genetic correlation (rA). The moderate rP for grain yield suggests that changes in rank will occur between a cultivar trial in a conventional system and a cultivar trial in an organic system. Moderate rP were also obtained for root lodging, stay green, and ear height whereas higher rP were obtained for grain moisture and plant height.


View this table:
[in this window]
[in a new window]

  		Table 4. Estimates of phenotypic correlation coefficients (rP) and genetic correlation coefficients (rA) in B73 x Mo17 recombinant inbred x LH295 testcrosses in conventional and organic production systems, and efficiency of indirect selection (CRorg/Rorg) in conventional system for performance in organic system.

 
In contrast to the moderate rP for grain yield, the rA between grain yield in the organic and conventional systems was high (rA = 0.84, Table 4). The high rA indicated that grain yield in the two production systems was influenced largely by the same set of genes, and that a differential response of the genotypes to the two production systems was largely absent. Genetic correlation estimates were also high (rA > 0.90) for grain moisture, plant height, and ear height. For these three traits, VTCxSys was not significant (not shown). On the other hand, rA was moderate for stay green and was nonsignificant for root lodging.

Selection in the conventional system, to improve performance in the organic system, will be more effective than selection in the organic system itself if the product of rA and the square root of h2 in the conventional system is greater than the square root of h2 in the organic system. The ratio of the correlated response to the direct response to selection in the organic system (CROrg/ROrg) was near 1.0 for grain yield (CROrg/ROrg = 0.94) and moisture (CROrg/ROrg = 0.95) (Table 4). In other words, indirect selection in the conventional system for grain yield and moisture will be nearly as effective as direct selection for grain yield and moisture in the organic system. CROrg/ROrg was also high for plant height (0.91) and ear height (0.83). For root lodging and stay green, however, CROrg/ROrg was low, suggesting that indirect selection will not be efficient. The lower predicted efficiency of indirect selection for root lodging and stay green was mainly due to the lower heritability of these traits in the conventional system than in the organic system.

Assessing the Need for a Separate Breeding Program for Organic Corn
Two pieces of information from this study indicated that to improve grain yield and moisture, a corn breeding program for organic systems separate from current breeding programs for conventional, high-input systems may not be needed. First, the higher VTC, lower VError, and higher h2 estimates for grain yield in the conventional system than in the organic system justifies indirect selection because genetic differences for yield may be better expressed in the conventional system. Second, the estimates of rA and of CROrg/ROrg indicated that improvement of corn grain yield and moisture under organic conditions can be achieved largely as a correlated response to selection in conventional systems. For other traits, however, the results of this study do indicate a need to conduct testing and selection under organic conditions, particularly for stay green and root lodging. Due to higher VTC and h2 estimates in the organic system than in the conventional system, selection for lower root lodging and increased stay green, as well as desired plant height and ear height, may be better performed under organic conditions.

Many of the previous studies utilizing the concepts of genetic correlation and correlated response to selection have found the need to at least include the stressed and low-input environments or to establish a separate breeding program for various types of low-input, stressed or low-productivity environments in oat (Avena sativa L.; Atlin and Frey, 1989, 1990), wheat (Triticum aestivum L.; Ud-din et al., 1992; Hill et al., 1999; Brancourt- Hulmel et al., 2005), barley (Hordeum vulgare L.; Ceccarelli et al., 1992; Ceccarelli, 1994) and corn (Presterl et al., 2003). In contrast, other studies indicated that indirect selection in high-N environments may be equally effective as direct selection for yield in low-N environments in oat (Atlin and Frey, 1989) and in corn (Brun and Dudley, 1989). In corn, indirect selection in high N conditions for performance in low-N environment may be less efficient than direct selection at low-N when the N stress exceeds a threshold level (Banziger et al., 1997). In this study, the high soil-fertility levels in both the organic and conventional systems (Table 1) may have led to the high rA and CROrg/ROrg values for yield.

Production practices for organic corn differ widely across the U.S. Corn Belt (Thomison et al., 2007), and the results from our study may not directly apply to organic systems that differ substantially from those used in this study (Table 1). Our comparisons were facilitated by having both conventional and organic sites at the same locations (Lamberton and Waseca, MN). We were unable, however, to repeat our experiments during a second year, and our results may not directly apply to environments that, unlike Waseca and Lamberton, MN, in 2006, did not have favorable growing conditions. Differences in planting date were confounded with production systems. Weed control is the most frequently cited concern by organic-corn producers (Thomison et al., 2007), and delayed planting allows mechanical removal of early flushes of weeds. The later planting dates for the organic systems in the current study were therefore consistent with usual, if not inherent, practices in organic-corn production (Porter et al., 2003).

Genetic variance, heritability, and genetic correlation estimates are influenced by gene frequencies and are population-specific (Falconer and Mackay, 1996). Therefore, the results of this study strictly apply only to the B73 x Mo17 population used and not necessarily to other populations. However, B73 was derived largely from Reid Yellow Dent, which represents around 51% of the background of U.S. hybrid corn, whereas Mo17 was derived from Lancaster Sure Crop, which represents some 13% of the U.S. hybrid corn background (Troyer, 2004). Both B73 and Mo17 have been extensively used as parents in developing elite maize inbreds in the U.S. Corn Belt (MBS Genetics, 2003). Therefore, the results of this study may be generally applicable to much of the current U.S. Corn Belt hybrid germplasm.

One may also argue that the results of this study might have limited applicability because the germplasm used was not bred specifically for organic systems. The parents of the recombinant inbreds in this study were developed in the late 1950s to 1960s: Mo17 was released in Missouri in 1964, and B73 was released in Iowa in 1972 (Troyer, 2004). Less N fertilizer and herbicides were used in the 1950s and 1960s than in the 2000s. In 1965, the average rates of fertilizer applied were 84-56-54 kg N-P-K ha–1 in the United States and 84-44-41 kg N-P-K ha–1 in Missouri (Vroomen, 1989). In 1970, the average rates of fertilizer applied were 125–74–77 kg N-P-K ha–1 in the United States and 120–75–71 kg N-P-K ha–1 in Iowa. In contrast, the rates of N fertilizer for the conventional systems in this study were 157 kg N ha–1 in Waseca and 168 kg N ha–1 in Lamberton (Table 1). Pesticide use then was not as widespread as today: amounts of total pesticides (active ingredients) applied to corn were 0.70 kg ha–1 (62% herbicides) in 1964 and 1.53 kg ha–1 (80% herbicides) in 1971 (Lin et al., 1995). In this study, rates of herbicide (active ingredients) application in the conventional systems were approximately 3.32 kg ha–1 in Waseca and 3.05 kg ha–1 in Lamberton (Table 1). These comparisons suggest that B73 and Mo17 were developed under production systems that were low-input compared with today's high-input conventional production systems, such as those used in this study. Thus, the materials used in this study may be considered as neutral in terms of production systems: B73 and Mo17 were not developed under conditions resembling either organic production systems or in current high-input conventional production systems. Predicted genetic gains or response to selection in organic and conventional systems for the germplasm in our study may therefore be comparable to the amount of selection response using other germplasm selected for lower-input production systems.

On the other hand, there may be valid justifications for a separate breeding program for organic corn production. For example, some organic producers may prefer to grow open-pollinated cultivars rather than hybrid cultivars. Also, an increasing number of conventional hybrids are genetically modified; in 2004, 63% of the corn cultivars grown in Minnesota and 47% in the United States were genetically modified (USDA-NASS, 2006). Because organic agriculture rejects the use of genetically modified organisms, organic farmers may not be able to depend on conventional breeding programs as sources of cultivars (Lammerts van Bueren and Verhoog, 2006) unless breeders remove transgenes from conventional inbreds by backcrossing. Nevertheless, for nontransgenic corn, the results of this study suggest that a separate breeding program for organic corn may not be needed: inbreds and hybrids may be developed under conventional systems and later tested or screened under organic conditions.


   ACKNOWLEDGMENTS
 
R.E. Lorenzana was supported by a Troyer/Darwin graduate fellowship at the University of Minnesota. We thank Tom Hoverstad, Steve Quiring, and Eric Ristau for their technical assistance.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication August 20, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
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 Lorenzana, R. E.
Right arrow Articles by Bernardo, R.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Lorenzana, R. E.
Right arrow Articles by Bernardo, R.
Agricola
Right arrow Articles by Lorenzana, R. E.
Right arrow Articles by Bernardo, R.
Related Collections
Right arrow Maize
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