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

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Welcker, C. Articles by Horst, W. J. Search for Related Content PubMed Articles by Welcker, C. Articles by Horst, W. J. Agricola Articles by Welcker, C. Articles by Horst, W. J. Related Collections Crop Genetics Maize Plant and Soil Interactions

CROP BREEDING, GENETICS & CYTOLOGY

Heterosis and Combining Ability for Maize Adaptation to Tropical Acid Soils

Implications for Future Breeding Strategies

^a INRA, Centre de Montpellier, UMR LEPSE, 2 Place Viala, 34000 Montpellier cedex1, France
^b INRA Centre Antilles Guyane, URPV, Domaine Duclos, Prise d'Eau, 97170, Petit-Bourg, Guadeloupe, France
^c IRAD, Maize Program, PO BOX 2067, Yaounde, Cameroon
^d CIMMYT, Programa de Maiz-Suramerica, CIAT, AA 6713, Cali, Colombia
^e EMBRAPA, CNPMS, Caixa postal 151, 35701-970, Sete Lagoas, MG, Brazil
^f CORPOICA, La Libertad, Villavicencio, Colombia
^g INRA-UPS-INAPG, Station de Génétique Végétale, Ferme du Moulon, 91190, Gif-sur-Yvette, France
^h UHANN, University of Hannover, Institute for Plant Nutrition, Herrenhaeuser Strasse 2, D-30419, Hannover, Germany

^* Corresponding author (welcker{at}ensam.inra.fr)

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Soil acidity reduces maize (Zea mays L.) yields by up to 70% on 8 million hectares in developing countries. Several breeding programs have produced populations better adapted to these conditions. The objectives of this study were to evaluate these populations for both per se cultivation and the development of new breeding germplasm. To do so, we generated a diallel cross design, which included six acid soil-tolerant and five susceptible populations with high yield potential or tolerance to other stresses. Populations and crosses were evaluated in five environments, on acidic Al-toxic soils and in comparable limed soils in Guadeloupe, Cameroon, and Colombia. Soil acidity decreased grain yield by 46 to 73%, depending on the location and year. Significant genotype x soil condition interactions were observed for grain yield. Mid-parent heterosis for yield was significantly higher in acid soils (32%) than in nonacid soils (20%). This suggests that the development of variety crosses between acid soil-tolerant populations could be used to increase maize yields in acid-soil cropping systems. The observed high general combining ability (GCA) for yield variation of the crosses in acid soil and its close relationship to per se performance suggest that parental populations of variety crosses could be efficiently screened on the basis of per se performance in acid soil.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
ACID SOILS cover approximately 3950 million hectares, corresponding to about 30% of the total ice-free land area on the Earth. Acid soils are found throughout the world, with 41% in the Americas, 26% in Asia, 17% in Africa, 10% in Europe, and 6% in Australia and New Zealand (Von Uexkull and Mutert, 1995). In the tropics, more than 8 million hectares of acid soils are planted with maize. This crop is not acid soil-tolerant (Pandey and Gardner, 1992), but increases in the demand for cereals in developing countries have led to an increase in maize acreage on acid soils.

Acidic soils have a low pH because the materials from which they are derived have a low basic cation (Ca, Mg, K, and Na) content and/or because these elements have been removed from the soil by leaching or via harvested crops (Granados et al., 1993). Acidic soils, therefore, generally have a low pH, contain toxic levels of Al and Mn, and are deficient in Ca, Mg, P, K, and Mo. These characteristics limit the fertility of acid soils and inhibit root development, leading to low water and nutrient uptake and low maize yields (Duque-Vargas et al., 1994).

Soil amendments (the application of lime and fertilizers) have been used to bring acid soils under agricultural production. However, such solutions may not be environmentally friendly, have only a temporary effect, and are too expensive for poor farmers in developing countries (Thé et al., 2001). The use of acid soil-tolerant maize cultivars provides an environmentally friendly, inexpensive, and permanent solution, contributing to sustainable crop production on acid soils (Granados et al., 1993).

Considerable genetic variation in acid-soil tolerance has been reported in maize. Early studies demonstrated qualitative inheritance for Al resistance (Rhue et al., 1978; Miranda Filho et al., 1984). Quantitative inheritance was subsequently demonstrated for the adaptation of maize to soil acidity (Lima et al., 1992; Duque-Vargas et al., 1994; Pandey et al., 1994; Borrero et al., 1995; Salazar et al., 1997). Recurrent selection has increased maize yields in acid soils, although major genotype x environment interactions complicate selection (Duque-Vargas et al., 1994; Borrero et al., 1995). Granados et al. (1993) reported increases in yield of 2% per cycle after 14 cycles of modified ear-to-row selection and 14% per cycle after two cycles of full-sib selection. Heterosis has been reported for yield in acidic conditions (Lima et al., 1992; Pandey et al., 1994; Salazar et al., 1997; Ceballos et al., 1998). This led to the extensive selection of inbred lines for hybrid development (Narro et al., 2000).

Significant progress has been made in the development of acid soil-tolerant and Al-resistant germplasm in recent years (Bahia Filho et al., 1997; Granados et al., 1993). In addition, several cultivars, such as the open-pollinated variety (OPV) ‘Sikuani,’ derived from SA3 for the Colombian llanos (Corpoica, 1998), ‘ATPY’ for the African humid forest zone (Thé et al., 1997), and CMS 36 for the Brazilian cerrado (Bahia Filho et al., 1997), have been selected for adaptation to specific environments. Little is currently known about the combining ability of acid soil-tolerant germplasm. Breeders from Latin America, Africa, and Europe, within the framework of a joint program (http://www.maizeforacidsoils.com/; verified 19 July 2005), have exchanged advanced genetic material and evaluated it in a wide range of acid soil environments. We used diallel crosses (i) to assess the performance of source germplasm in various target environments, (ii) to determine combining ability of tropical maize populations differing in adaptation to acid soil environments, and (iii) to study the genetic basis of adaptation and related traits. Improving our knowledge of the breeding value of acid soil-tolerant populations and the magnitude of heterosis for yield would help breeders to identify suitable germplasm and to define more appropriate methods for the development of well-adapted varieties with high yielding capacity on acid soils for the resource-poor farmers of the tropics.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Eleven maize populations with different genetic backgrounds were used as parents in a diallel cross study (Table 1). These populations included six segregating maize populations in which tolerance to soil acidity had already been increased by recurrent selection (ATPY, ATPW, SA7, SA6, SA4, SA3) and five populations with high yielding capacity or tolerance to other stresses such as drought and insect damage (Tuxpeño, BR106, Spectral, Kristal, CMS9213). Plant-by-plant crosses (100 plants per population per cross) were made in the winter of 1999 in Cameroon for the first year of evaluation and in the winter of 2000 for the second year of evaluation. For each production year, about 60 ears were bulked for each reciprocal cross, assuming an absence of significant reciprocal effects (Salazar et al., 1997).

Plant vigor was assessed at the six-leaf stage (from 1 = vigorous and healthy plant to 9 = very poor growth and visible nutrient disorder), and plant height was measured 15 d after silking (cm from ground level to the base of the tassel). We also recorded the number of days from planting to 50% anthesis (50% of plants in the plot having anthers), number of days between 50% silking and 50% anthesis (anthesis to silking interval or ASI), number of ears per plant with at least one kernel, and grain yield (kg/m² converted to Mg ha^–1).

Analyses of variance were first performed by the GLM procedure of SAS (SAS Institute, 1989). Environments, soil conditions, and genotypes (populations or crosses) were considered fixed effects, whereas replications within environments were considered random effects. Further genetic modeling of cross effects was conducted on the average performance of the cross in a given trial (environment x soil conditions). The error variance of the average performance in a given trial (t), ²_e,t was estimated as the plot-level error-variance of the trial (by the VARCOMP procedure of SAS, SAS Institute, 1989) divided by the number of replicates of the trial. We then computed the pooled error variance, ²_e, of the trials included in a given analysis as the average of corresponding ²_e,t values. We performed diallel cross analysis for all variables, according to Griffing's (1956) method 4 (excluding parents and reciprocal F₁s) model 1 (fixed model), to estimate general combining ability (GCA) and specific combining ability (SCA). Population performance per se was not included in these analyses so that we could focus on evaluation of combining ability with unrelated material. Data were analyzed by a program (Andréau, Epinat le Signor and Charcosset, personal communication) developed in the IML (interactive matrix language) procedure of SAS (SAS Institute, 1990). This program can be used to analyze unbalanced diallel crosses in which reciprocals and/or selfs are absent or are present in unequal numbers. The contributions of GCA and SCA effects to variations in cross performance were defined respectively as _GCA = g²_i and _SCA = s²_ij, where p is the number of parents, g_i is the general combining ability of parent i, s_ij is the specific ability of cross ij. Note that these parameters are related to those of Table 4.7 in Hallauer and Miranda (1988) as _SCA = K²_SCA and _GCA = K²_GCA. They were estimated by the Method of Moments, based on mean square (MS) expectations (Singh, 1973) as _GCA = and _SCA = , where p is the number of parents and h the number of crosses, MS_GCA and MS_SCA are the mean squares of GCA and SCA effects, respectively, ²_e is the pooled experimental error described above, and l is the number of trials. The relative contribution of SCA to total genetic variation was estimated as _SCA/(_GCA + _SCA).

Heterosis for grain yield was calculated with respect to the mean values of the parents used in the cross considered. Pearson's correlation coefficients (using the CORR procedure of SAS; SAS Institute, 1989) were also calculated to determine relationships between traits and between performances for a given trait evaluated in acid and nonacid soil conditions.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Effect of Soil Acidity, and Performance of Populations, and Their Crosses
Grain yield losses due to soil acidity ranged from 46 to 73% (Table 2). The highest yield losses were observed in Guadeloupe in 1999, where P fertilizer was not applied to the acid plot. The lowest yield losses were observed in Colombia in 2000 probably because of residual Al toxicity in the soil at Villaviciencio, which was used as the reference nonacid soil for the Colombian environment. Highly significant differences between acid and nonacid soil conditions were also observed for all secondary traits in each environment, with the exception of ASI and ears per plant in Colombia (data not shown).

The grain yields of the parental populations varied considerably in both acid and nonacid soils (Fig. 1 and Table 3). The parental populations in which acid-soil tolerance had been increased by recurrent selection largely outperformed the unselected populations in terms of yield (1.87 vs 1.28 Mg ha^–1; see Table 4). This reflects the progress made in improving the adaptation of maize populations to acid soils in the last few decades. Tolerant populations from Colombia (SA) and from Cameroon (ATP) performed best at their respective selection sites (data not shown). In the absence of P application to acid soil (Guadeloupe, 1999), ATPY yields declined dramatically, whereas SA7 yields remained acceptable (2.31 Mg ha^–1), indicating differences in adaptation between tolerant germplasm.

View larger version (18K): [in this window] [in a new window]   Fig. 1. Relationship between mean grain yields of 11 parental maize populations and their 55 crosses in acid and nonacid soil environments. The correlation between grain yield in acid and nonacid soils is 0.32 (P < 0.05). Note, crosses could be classified as "acid-soil tolerant with high yielding capacity in nonacid soil" (quadrant 1), "acid-soil tolerant with low-yielding capacity" (2), "acid-soil susceptible with low-yielding capacity" (3), and "acid-soil susceptible with high yielding capacity" (4). (Gray-screened circle) parental populations, (filled diamond) crosses.

In acid soil conditions, yield was highly significantly correlated with plant vigor at the six-leaf stage, indicating that faster early growth led to higher yields (Table 5). Highly significant negative correlations were found between days to anthesis/ASI and ears per plant. Relationships between grain yield and reproductive traits (anthesis, ASI, and ears per plant) indicated that earliness and the maintenance of synchrony in male and female flowering made a major contribution to higher yields on acid soils. In nonacid soil conditions, the phenotypic correlation coefficients for the relationship between yield and related traits were similar in sign to those observed in acid soil conditions, but these relationships were generally weaker (Table 5).

Diallel Analysis
When considering both acid and nonacid soil conditions in all five environments, we found that soil condition, GCA effects, and their interactions with soil condition had significant effects on yield (Table 6). In contrast, SCA effects averaged across soil types were significant only for anthesis and ASI, and SCA x soil conditions interactions were never significant. If we considered acid and nonacid soil conditions in each individual environment (data not shown), GCA effects were significant (P = 0.05) for all traits, except in Cameroon for ASI in 1999 and for ears per plant in 2000, whereas SCA effects were seldom significant. The GCA x soil condition interactions were significant in most of the environments tested, indicating that crosses between parental populations behaved differently according to the environmental conditions.

View larger version (18K): [in this window] [in a new window]   Fig. 2. Observed grain yields (Mg ha–1) across acid-soil conditions of the 55 F1 diallel crosses, and their predicted grain yields (Mg ha–1) calculated for each cross from the GCA effect of two parents and the over-all mean. Note that deviation of the points from the 1:1 line was due in part to specific combining ability (SCA) effects. (R2 = 0.493).

The GCA effect, which determines the average performance of a parent in a series of crosses, was calculated for each parental population (Table 4). In acid soil conditions, the maximum favorable GCA effects were observed for the acid soil-tolerant populations SA4 and SA7. These populations had the largest favorable GCA effects on yield in each tested environment (data not shown). Positive GCA effects on yield in acid soil conditions were also observed for the tolerant populations ATPW and SA3. Favorable GCA effects for plant vigor and plant height were observed for ATPW. The population ATPY, the most Al-resistant parent, had low GCA, except in Cameroon, in the most Al-toxic soil environment. The GCA effects on flowering (anthesis and ASI) were generally negative for the tolerant parents and positive for the susceptible parents, indicating that early-flowering genotypes had higher combining abilities for yield in acid soil environments. In nonacid soil conditions, BR106 displayed the largest GCA for yield (0.38 Mg ha^–1) and SA7 the second largest GCA (0.35 Mg ha^–1).

Correlations between the per se performance of parental populations and their GCA effects in acid soil conditions were strong and significant for all traits (Table 4), indicating that suitable parental populations for hybrid development could be identified on the basis of their per se performance. The tolerant populations SA4 and SA7 can, therefore, be considered superior to the other populations.

The three best hybrids were associated with the three largest positive SCA effects (>0.44) (Table 8). These hybrids were produced exclusively by crosses between acid soil-tolerant populations (SA7, SA4, ATPW, and SA6) with positive GCA effects.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

Consistent with previous reports (Pandey et al., 1994; Ceballos et al., 1998), we observed a significant positive phenotypic correlation between yield in acid and nonacid soil conditions (Fig. 1, Table 5). Bänziger et al. (1997) reported that the genetic correlation between yield in N stress and optimal environments tends to decrease as stress intensity increases. In this study, the correlation was low (0.32) if we considered mean performance in acid and nonacid soils (Fig. 1). Therefore, yield evaluation for target acid-soil environments would probably be most accurate if based on data from acid soils only (Duque-Vargas et al., 1994). Consistent with the weak correlation between yield in acid and nonacid soil conditions, interactions between GCA and soil conditions for yield were significant. This suggests that selection in an optimal environment is unlikely to be the most effective way of identifying superior genotypes for stress environments. Thus, selection for adaptation to acid soils requires the use of managed soil conditions and/or physiological tests (Schaffert et al., 2000; Collet et al., 2002; Eticha et al., 2005; Sierra et al., 2005).

Soil acidity affected most traits related to plant growth and development, including plant vigor at an early stage (–70% on average), male flowering (+9 d), plant height (–36%), and traits related to reproduction, such as ASI (+2.5 d) and number of ears per plant (–21%). Sierra et al. (2003) and Bonhomme et al. (2004) recently showed that the combination of Al toxicity and P deficiency has a severe effect on the photosynthetic system, reducing individual leaf area, delaying the appearance of leaves, and leaf development, thereby also delaying male flowering. Aluminum, the most important toxic factor in acid soils, primarily inhibits root elongation, which has been the basis for selection in screening cultivars for acid soil tolerance (Foy et al., 1993). Restriction of root growth, particularly in the subsoil, will limit maize yields by impeding uptake of nutrients (Sierra et al., 2003) and water (Goldman et al., 1989), increasing the sensitivity of the crop to drought (Edmeades et al., 1993) and N-deficiency stresses (Bänziger and Lafitte, 1997). Since tropical maize production generally relies on rainfall and low fertilizer inputs, maize breeders, therefore, should not concentrate only on Al resistance but also try to develop tolerance to multiple stresses in maize cultivars for acid soil environments.

Importance of Heterosis
In nonacid soil conditions, mid-parent heterosis was 20% (Table 3), which is close to or slightly higher than the values reported previously for crosses between noninbred parental materials (Miranda Filho et al., 1984; Crossa et al., 1990; Vasal et al., 1992; Spaner et al., 1996; Hallauer and Miranda Filho, 1998; Kim et al., 1999; Parentoni et al., 2001). Interestingly, mid-parent heterosis for yield was higher (32%) on acid soils. This stronger heterosis under stress conditions than under optimal conditions is consistent with the findings of Betran et al. (2003).

Supplying superior variety crosses to farmers may be a highly valuable strategy. We found no relationship between heterosis and geographical origin of the parental populations or between heterosis and tolerance to soil acidity of the parental populations. However, in general, tolerant x tolerant crosses (2.42 Mg ha^–1) outyielded susceptible x tolerant crosses (2.07 Mg ha^–1) and susceptible x susceptible crosses (1.81 Mg ha^–1), indicating polygenic inheritance for this trait and the better adaptation of tolerant x tolerant combinations (Ceballos et al., 1998).

Genetic Basis of Adaptation to Acid Soils
Our study confirmed that GCA has a strong effect on the inheritance of adaptation to acidic, Al-toxic soils in maize. This was also observed by Duque-Vargas et al. (1994) for the tolerant SA3 germplasm and by Magnavaca et al. (1987) and Lima et al. (1992) for Al resistance in terms of root elongation in nutrient solutions containing toxic concentrations of Al. Eticha et al. (2005) evaluated Al-induced callose formation in root apices in the same diallel set and concluded that Al resistance is mediated largely by additive genes. Although SCA ap- peared to be of only minor importance, it made a greater relative contribution to variations in the performance of crosses in acid soils (0.28) than in nonacid soils (0.03). These results are consistent with the differences in heterosis reported above and with previous studies in related genetic backgrounds (Salazar et al., 1997; Ceballos et al., 1998).

The strong correlation between per se performance and GCA in acid conditions indicates that (i) parental populations for use in the production of variety crosses can be efficiently screened on the basis of per se performance and (ii) recurrent selection for the development of open-pollinated varieties is also efficient for the development of parents of acid soil-tolerant hybrids (Table 4).

The GCA x environment interaction was the second largest source of variation (based on mean squares) after GCA, for yield in acid soil conditions (Duque-Vargas et al., 1994; Pandey et al., 1994). Thus, the selection of tolerant material would be more effective if based on performance across a range of acid soil environments. Lower GCA x environment interactions for early traits (such as plant vigor at the six-leaf stage) and later traits (such as ears per plant) may indicate that selection would be more efficient if based on these traits. The positive correlation between these traits and grain yield in terms of GCA (0.52 and 0.71, respectively) suggests that they would be useful as selection traits. Estimation of genetic correlations and heritabilities of these traits would be needed to accurately predict the response to indirect selection.

Identification of Suitable Germplasm for the Development of Acid-Tolerant Varieties
The parental populations SA4, SA7, ATPW, and SA3 had favorable GCA effects and performed well in acid soil environments, indicating a high level of adaptation to these target environments. The population SA3 has been shown to be potentially well-adapted to soils with a low P content (Gaume et al., 2001; Sierra et al., 2003), whereas ATW was selected specifically for Al resistance (Thé et al., 1997). The population ATPY did not have a strong GCA effect on Al-induced callose formation in Al-enriched nutrient solution in this diallel set (Eticha et al., 2005), which is consistent with the results presented here.

Adaptation to acid soils could be improved in several ways, including development of new improved open-pollinated varieties, variety crosses, and F1 hybrids. The development of new open-pollinated varieties could include recurrent selection in a pool composed of the white maize populations SA7 and ATPW and a second pool composed of the yellow maize populations SA4 and SA3. The population SA7 was the earliest flowering variety and had the lowest ASI, which is particularly interesting in environments in which drought may occur (Betran et al., 2003). Further improvement of this specificity could be obtained by crossing SA7 with TUXP, which had a highly favorable GCA for reducing ASI in acid soils. Other materials not evaluated in this study may also be considered. Parallel diallel studies, including the plant material tested here and other unrelated germplasm from South America, have shown that CMS36, developed from Cateto and ETO material for adaptation to the Cerrado in Brazil (Bahia Filho et al., 1997), displayed high GCA for yield in acid soil conditions (Pandey et al., 1994; Welcker and Parentoni, personal communication). Eticha et al. (2005) also reported that crossing with CMS36 increased Al resistance. The CMS36 population is probably from a different genetic background than that of the acid soil tolerant materials analyzed in this study (Parentoni et al., 2001). It is, therefore, a potential donor of favorable alleles (Schaffert et al., 2000), the utilization of which would increase the chances of long-term progress in breeding maize for adaptation to acid soils.

On the basis of the strong heterotic effect observed in acid soils, we consider the development of variety crosses to be a potentially valuable alternative to that of open pollinated varieties. Unlike F1 hybrids, the seeds of such hybrids can easily be produced locally in developing countries. Moreover, the clear advantage of the best crosses over the best population (50% on average, Table 3) provides resource-poor farmers with an opportunity to avoid having to regenerate seeds at each planting generation. Our results suggest that breeders in specific regions could deliver variety crosses for local uses. These crosses include the white dent variety type ATPW x SA7 for the humid forest agro-ecosystem, the yellow flint types SA4 x SA3 for the Colombian savanna, or KRIS x SA4 for Caribbean cropping systems on acid soils. In the five environments studied, the SA7 x SA4 cross differed from the others in giving high yields on both acid and nonacid soils (quadrant 1 in Fig. 1). This cross involved the two highest-yielding populations, SA7 and SA4, which are acid-soil tolerant. This finding illustrates the progress made by CIMMYT in improving the yields of population varieties for acid soils in the last decade (Narro et al., 2001). However, these two populations do not have individually sufficient residual variability for a long-term selection program (De Leon, personal communication). Thus, there exists a need to develop new populations for further breeding. Considering the promising perspectives offered by the development of variety cross hybrids, the choice of parental material for these populations could be made by evaluating their combining ability with both SA4 and SA7 (used as testers). Materials displaying a good combining ability with SA7 could also be intercrossed with SA4, whereas those that combined well with SA4 could be intercrossed with SA7.

Finally, a further step toward performance improvement in acid soil conditions could be achieved through the extraction of inbred lines from improved Al-tolerant populations, followed by the crossing of lines from different heterotic groups to produce superior hybrids as done by CIMMYT (Narro et al., 2000).

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Our results confirm that Al resistance makes a substantial contribution to adaptation to acid soils. A large proportion of maize crops are grown by small-scale farmers without irrigation and limited possibilities for other inputs. Therefore, low soil P availability and dry spells during the main growing period are major yield-limiting factors on many acid soils, making improvements in the efficiency of P use and drought tolerance equally important. Breeding for adaptation to acid soils should focus on tolerance to multiple stresses. Genetic variation that was observed suggests that recurrent selection will be effective for the development of new improved populations. The development of open pollinated varieties is still required as more than 80% of the maize planted in the tropics corresponds to this type of variety. However, the increase in heterosis with greater stress, and the contribution of GCA to yield variation in variety crosses, suggest that the development of variety crosses between tolerant populations could be a promising way of increasing the productivity and sustainability of maize-based farming systems on tropical acid soils.

	ACKNOWLEDGMENTS

 
We would like to thank B. L'Hôte, M. Al Rifaï, C. Lorenz (graduate students from France and Germany), J.P. Cinna, P. Renac (technical assistants in Guadeloupe), H. Calba (CIRAD), B. Jaillard (INRA) (soil characterization) for their collaboration in the various phases of this study. We would also like to thank Christine Epinat-Le Signor (INRA) for her contribution to the development of the SAS IML diallel analysis program used for this study. We also thank A. Gallais (geneticist), R Bonhomme (agronomist), and anonymous Crop Science reviewers for their constructive review of an earlier draft of this manuscript. This collaborative work was supported by the European Union, Science-Research-Development, International Cooperation with Developing Countries (IC18 CT96 0063 and ICA4 CT 2000 30017).

Received for publication October 14, 2004.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 

• Bahia Filho, A.F.C., R. Magnavaca, R.E. Schaffert, and V.M.C. Alves. 1997. Identification, utilization, and economic impact of maize germplasm tolerant to low levels of phosphorus and toxic levels of exchangeable aluminium in Brazilian soils. p. 59–70. In A.C. Monitz et al. (ed.) Plant-soil interactions at low pH. Brazilian Soil Science Society, Campinas, SP, Brazil.
• Bänziger, M., and H.R. Lafitte. 1997. Breeding tropical maize for low N environments: II The value of secondary traits for improving selection gain under low N. Crop Sci. 37:1110–1117.[Abstract/Free Full Text]
• Bänziger, M., F.J. Betran, and H.R. Lafitte. 1997. Breeding tropical maize for low N environments: I Spillover effects from high N selection to low N target environments. Crop Sci. 37:1103–1109.[Abstract/Free Full Text]
• Betran, F.J., D. Beck, M. Bänziger, and G.O. Edmeades. 2003. Genetic analysis of inbred and hybrid grain yield under stress and nonstress environments in tropical maize. Crop Sci. 43:807–817.[Abstract/Free Full Text]
• Bonhomme, R., M. Al Rifaï, J. Sierra, J. Félicité, and C. Welcker. 2004. The size and the dynamic of the maize leaf system are profoundly modified in acid soils. p. 369–370. In S.E. Jacobsen et al. (ed.) Proceedings of the VII European Society of Agronomy Congress, European Agriculture in Global Context, Denmark.
• Borrero, J.C., S. Pandey, H. Ceballos, R. Magnavaca, and A.F.C. Bahia Filho. 1995. Genetic variances for tolerance to soil acidity in a tropical maize population. Maydica 40:283–288.
• Ceballos, H., S. Pandey, L. Narro, and J.C. Perez-Velàzquez. 1998. Additive, dominant, and epistatic effects for maize grain yield in acid and non-acid soils. Theor. Appl. Genet. 96:662–668.[CrossRef]
• Collet, L., C. De Leon, M. Kollmeier, N. Schmohl, and W. Horst. 2002. Assessment of aluminium sensitivity of maize cultivars using roots of intact plants and excised root tips. J. Plant Nutr. Soil Sci. 165:357–365.[CrossRef]
• Corpoica. 1998. Acid tolerant maize: Promoting sustainable farming on acid savannas. Release of ‘Sikuani’ by Colombian Agricultural Research Corporation in collaboration with CIAT and CIMMYT.
• Crossa, J., S.K. Vasal, and D.L. Beck. 1990. Combining ability estimates of CIMMYT's tropical late yellow maize germplasm. Maydica 35:273–278.
• Duque-Vargas, J., S. Pandey, G. Granados, H. Ceballos, and E. Knapp. 1994. Inheritance of tolerance to soil acidity in tropical maize. Crop Sci. 34:50–54.[Abstract/Free Full Text]
• Edmeades, G.O., J. Bolaños, M. Hernadez, and S. Bello. 1993. Causes for silk delay in a lowland tropical maize population. Crop Sci. 33:1029–1035.[Abstract/Free Full Text]
• Eticha, D., C. Thé, C. Welcker, and L. Narro. A. Staß, and W. Horst. 2005. Aluminium-induced callose formation in root apices: Inheritance and selection trait for adaptation of tropical maize to acid soils. Field Crop Res. 93:252–263.[CrossRef]
• Foy, C.D., T.E. Carter, Jr., J.A. Duke, and T.E. Devine. 1993. Correlation of shoot and root growth and its role in selecting for Al tolerance in soybean. J. Plant Nutr. 16:305–325.
• Gaume, A., F. Mächler, and E. Frossard. 2001. Aluminium resistance in two cultivars of Zea mays L.: root exudation of organic acids and influence of phosphorus nutrition. Plant Soil 234:73–81.[CrossRef]
• Goldman, I.L., T.E. Carter, Jr., and R.P. Patterson. 1989. A detrimental interaction of subsoil aluminium and drought stress on the leaf water status of soybean. Agron. J. 81:461–463.[Abstract/Free Full Text]
• Granados, G., S. Pandey, and H. Ceballos. 1993. Response to selection for tolerance to acid soils in a tropical maize population. Crop Sci. 33:936–940.[Abstract/Free Full Text]
• Griffing, B. 1956. Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci. 9:463–493.
• Hallauer, A.R., and J.B. Miranda Filho. 1988. Quantitative Genetics in Maize Breeding. 2nd ed. Iowa State Univ. Press, Ames, IA.
• Kim, S.K., S.O. Ajala, and J.L. Brewbaker. 1999. Combining ability of tropical maize germplasm in west Africa. III. Tropical maize inbreds. Maydica 44:285–291.[ISI]
• Lima, M., P.R. Furlani, and J.B. Miranda Filho. 1992. Divergent selection for aluminium tolerance in a maize (Zea mays L.) population. Maydica 37:123–132.
• Magnavaca, R., C. Gardner, and R. Clark. 1987. Comparison of maize populations for aluminium tolerance in nutrient solution, p. 189–199. In H. Gabelman and B.C. Loughman (ed.) Genetic aspects of plant mineral nutrition. Martinus Nijhoff Publ, Dordrecht, Netherlands.
• Miranda Filho, L.T., P.R. Furlani, L.E.C. Miranda Filho, and E. Sawazaki. 1984. Genetics of environmental resistance and super-genes: Latent aluminium tolerance. Maize Genet. Coop. Newslet. 58:46–48.
• Narro, L.A., J.C. Pérez, S. Pandey, J. Crossa, F. Salazar, M.P. Arias, and J. Franco. 2000. Diallel and triallel analysis in acid soil tolerant maize (Zea mays L.) population. Maydica 45:301–308.
• Narro, L.A., S. Pandey, C. De Leon, F. Salazar, and M.P. Arias. 2001. Implications of soil-acidity tolerant maize cultivars to increase production in developing countries. p. 447–463. In N. Ae et al. (ed.) Plant nutrient acquisition: New perspectives. NIAES series 4. Springer Verlag, Japan.
• Pandey, S., and C.O. Gardner. 1992. Recurrent selection for population, variety and hybrid improvment in tropical maize. Adv. Agron. 48:1–87.
• Pandey, S., H. Ceballos, R. Magnavaca, A.F.C. Bahia Filho, J. Duque-Vargas, and L.E. Vinasco. 1994. Genetics of tolerance to soil acidity in tropical maize. Crop Sci. 34:1511–1514.[Abstract/Free Full Text]
• Parentoni, S.N., J.V. Magalhaes, C.A.P. Pacheco, M.X. Santos, T. Abadie, E.E.G. Gama, P.E.O. Guimaraes, W.F. Meirelles, M.A. Lopes, M.J.V. Vasconcelos, and E. Paiva. 2001. Heterotic groups based on yield-specific combining ability data and phylogenetic relationship determined by RAPD markers for 28 tropical maize open pollinated varieties. Euphytica 121:197–208.[CrossRef]
• Rhue, R.D., C.O. Grogan, E.W. Stockmeyer, and H.L. Everett. 1978. Genetic control of aluminium tolerance in corn. Crop Sci. 18:1063–1067.[Abstract/Free Full Text]
• Salazar, F.S., S. Pandey, L. Narro, J.C. Perez, H. Ceballos, S.N. Parentoni, and A.F.C. Bahia Filho. 1997. Diallel analysis of acid-soil tolerant and intolerant tropical maize populations. Crop Sci. 37:1457–1462.[Abstract/Free Full Text]
• Sanchez, P.A., and J.G. Salinas. 1981. Low input technology for managing oxisols and utisols in tropical America. Adv. Agron. 34:279–405.
• SAS Institute. 1989. SAS/STAT guide for personal computers. Version 6. SAS Institute Inc., Cary, NC.
• SAS Institute. 1990. SAS/IML Software: Usage and reference. Version 6. lst ed. SAS Institute Inc., Cary, NC.
• Schaffert, R.E., V.M.C. Alves, S.N. Parentoni, and K.G. Raghothama. 2000. Genetic control of phosphorus uptake and utilization efficiency in maize and sorghum under marginal soil conditions, p. 79–85. In J.M Ribaut and D. Poland (ed.) Molecular approaches for the genetic improvement of Cereals for stable production in water-limited environments: A strategic planning workshop held at CIMMYT, El Batan, Mexico. 21–25 June 1999. CIMMYT, Mexico D.F.
• Sierra, J., C. Noël, L. Dufour, H. Ozier-Lafontaine, L. Desfontaine, and C. Welcker. 2003. Mineral nutrition and growth of tropical maize as affected by soil acidity. Plant Soil 252:215–226.[CrossRef]
• Sierra, J., H. Ozier-Lafontaine, L. Dufour, A. Meunier, R. Bonhomme, and C. Welcker. 2005. Nutrient and assimilate partitioning in two tropical maize cultivars in relation to their tolerance to soil acidity. Field Crop Res. (in press).
• Singh, D. 1973. Diallel analysis for combining ability over several environments-II. Indian J. Genet. Plant Breed. 33:469–481.
• Spaner, D., R.A.I. Brathwaite, and D.E. Mather. 1996. Diallel study of open-pollinated maize varieties in Trinidad. Euphytica 90:65–72.
• Thé, C., W.J. Horst, H. Calba, C. Welcker, and C. Zonkeng. 1997. Identification and development of maize genotypes adapted to acid-soil of the tropics, p. 65. In A.C. Moniz et al. (ed.) Plant-Soil Interactions at low pH: Sustainable agriculture and forestry production. Proc. 4th international symposium on plant-soil interactions at low pH, Belo Horizonte, MG, Brazil. 17–24 March 1996. Brazilian Soil Science Society, Campinas, SP, Brazil.
• Thé, C., H. Calba, W.J. Horst, and C. Zonkeng. 2001. Three years performance of a tolerant and a susceptible maize cultivar on non-amended and amended acid soil. p. 984–985. In W.J. Horst et al. (ed.) Plant nutrition-food security and sustainability of agro-ecosystems through basic and applied research. Hannover, 2000. Kluwer Academic Publishers, Dordrecht, the Netherlands.
• Vasal, S.K., G. Srinivasan, J. Crossa, and D.L. Beck. 1992. Heterosis and combining ability of CIMMYT's subtropical and temperate early-maturity maize germplasm. Crop Sci. 32:884–890.[Abstract/Free Full Text]
• Von Uexkull, H.R., and E. Mutert. 1995. Global extent, development and economic impact of acid soils. p. 5–19. In R.A. Date et al. (ed.) Plant-soil interaction at low pH: Principles and management. Kluwer Academic Publishers, Dordrecht, the Netherlands.

This article has been cited by other articles:

				C Welcker, B Boussuge, C Bencivenni, J-M Ribaut, and F Tardieu Are source and sink strengths genetically linked in maize plants subjected to water deficit? A QTL study of the responses of leaf growth and of Anthesis-Silking Interval to water deficit J. Exp. Bot., January 1, 2007; 58(2): 339 - 349. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Welcker, C. Articles by Horst, W. J. Search for Related Content PubMed Articles by Welcker, C. Articles by Horst, W. J. Agricola Articles by Welcker, C. Articles by Horst, W. J. Related Collections Crop Genetics Maize Plant and Soil Interactions