SEROPÉDICA, RIO DE JANEIRO, BRAZIL
Universidade Federal Rural do Rio de Janeiro

LANCASTER, UNITED KINGDOM
Biological Sciences, Lancaster University

PIRACICABA, SAÕ PAULO, BRAZIL
Escola Superior de Agricultura Luiz de Queiroz — USP

Combining ability of maize lines with varied responses to applied nitrogen

— Medici, LO; Pereira, MB; Lea, PJ; Azevedo, RA

Previous studies with maize have shown varied degrees of correlation between line and hybrid performance under low-N application, ranging from weak to a fairly strong relationship. Our research group has found that lines identified as having a high response to N supply for grain yield exhibited high general combining ability (GCA) effects for this trait, while the lines with low response to N supply exhibited low GCA effects. The results obtained indicate that the maize lines could be tested under both high and low N applications, and those with the highest contrast for grain yield and also for secondary traits such as chlorophyll content could be selected to produce hybrids.

Lafitte and Edmeades (Maydica 40: 259, 1995) observed that the best S3 maize lines at low N applications were not those with best effects of GCA at this level. The lack of association between the grain yield of lines and the grain yield of their hybrids, was also observed by Krone and Lambert (Maydica 40: 211, 1995; Maydica 40: 217, 1995) and Balko and Russel (Agronomy J. 72:723, 1980). However, Zaidi et al. (Maydica 48: 221, 2003) observed that the relationship between the grain yield of mid-parents and hybrids was comparatively stronger under low-N application. They suggested that the performance of hybrid progenies under low-N can be predicted on the basis of advance generation elite inbred parents, with proven performance across a range of N applications.

To gain more insight into the association between lines and hybrids of maize, we have carried out a diallel study using previously identified maize lines, as described in Medici et al. (J. Agriculture Science, in press, 2004). The 15 possible hybrids of six S5 lines of maize (Zea mays L.), developed at the Universidade Federal Rural do Rio de Janeiro, were used. The lines 2, 3, 4 and 10 were derived from the commercial hybrid AG 311, produced by the Agroceres seed company, Brazil, whereas lines 5 and 6 originated from the commercial hybrid AG 302 of the same company. The contrasting responses to N supply of these lines are reported in Medici et al. (J. Plant Nutrition, in press, 2005), in which two experiments were carried out (Tables 1 and 2). In the first, the N supplied appeared to be in excess and only increased the grain yield in line 2, whilst the chlorophyll content was increased in all lines, except line 3. In the second experiment, there was a severe N deficiency that caused a reduction in grain yield in lines 2, 3, 5 and 10; chlorophyll content in all lines (data not shown); prolificacy in lines 2, 3, 4 and 5; grain N content in all lines except line 10; and total plant N in lines 2, 3, and 5. On the other hand, such a severe N deficiency caused an increase in hundred kernel weight in line 2 and anthesis-silking interval in lines 3 and 5. In these experiments, line 2 had the highest anthesis-silking interval and line 6 had the lowest value for this trait. For chlorophyll content, the lines 2 and 6 were identified as having high values, while line 4 showed a low value for this trait. For prolificacy, line 2 was identified as having a high value.

In both the experiments of Medici et al. (J. Plant Nutrition in press, 2005) the two controls used were obtained from Embrapa Milho e Sorgo Brazil (Tables 1 and 2). One control was the population Sintetico Elite NT, which had been selected for high yield on low N soils, and the other was the cultivar BR 106 which was more sensitive to N deficiency than the first control in the work of Santos et al. (Pesquisa Agropecuária Brasileira 33:55, 1998). In the first experiment of Medici et al. (J. Plant Nutrition, in press, 2005), the lines 13 and 17 were also used, which originated from the commercial hybrid AG 311 of Agroceres seed company, Brazil (Table 1).

The experimental design of the diallel study was a factorial of the 15 hybrids, by two levels of N supply (10 and 130 kg N ha-1 as urea) in a randomized complete block design with six replicates. The plots consisted of a one-row plot 5 m long, spaced at 0.9 m, at a within-row distance of 0.2 m between plants. There was no association between the behaviour of lines per se at the low N application and their GCA effects at this level (Table 3). Only lines 4 and 6 were identified as having no significant grain yield reduction due to N deficiency, and line 6 was also identified as having the highest grain yield at low N (Medici et al., J. Plant Nutrition, in press, 2005); however, these lines had the two lowest effects of GCA at low N for this trait.

The lines with a high grain yield response to N supply had the largest effects of GCA for this trait. Only line 2 showed a grain yield response to high N supply in the first experiment (Medici et al., J. Plant Nutrition, in press, 2005). This line had the highest GCA effect for grain yield at high N, and the third best effect at low N. Line 5 was also identified as having a large response to N supply and had the greatest GCA effect for grain yield at low N, and the second highest at high N. Krone and Lambert (Maydica 40:211, 1995; Maydica 40:217, 1995) showed that lines selected following the application of high N were better than those selected at low N for the production of hybrids with the greatest grain yield. Balko and Russel (Agronomy J. 72:723, 1980) reported that nine out of ten important parent lines of commercial maize hybrids had a high response of grain yield to N supply. Duvick (Maydica 37:69, 1982) also indicated that lines with a high N response to N supply would be good for use as parents of hybrids. In the past, the parents of American simple cross hybrids of maize were selected for tolerance to high N supply; the new hybrids have a greater grain yield at low and high N supply (Duvick, Genetic Contribuition to Yield Gains of Five Major Crop Plants, Special Publication 7. (Ed. W. Fehr), pp 15–47, 1984). Frova et al. (TAG 99:280, 1999) reported that a maize line sensitive to drought provided a large number of alleles for grain yield. These authors argued that the sensitivity of grain yield to drought could have a different genetic control than grain yield itself.

The lines identified as having a high response to N supply exhibited high GCA effects for this trait, while the lines with low response to N supply exhibited low GCA effects. This association indicates that the response of the line to N supply should be considered in any future maize breeding program to achieve acceptable hybrids for environments with high and low N.

Table 1. F Test for the nitrogen (N) effect in each line (2, 3, 4, 5, 6, 10, 13 and 17) and controls BR 106 (BR) and Sintético Elite NT (SE) and averages at high and low N of grain yield and chlorophyll content in experiment 1.

*, **, NS — Significant at the 0.05 and 0.01 levels, and not significant at 0.05 level respectively. Values within a row followed by the same letter do not differ significantly (p < 0.05) according to the Tukey test.

Table 2. F Test for the nitrogen (N) effect in each line (2, 3, 4, 5, 6 and 10) and controls BR 106 (BR) and Sintético Elite NT (SE) and averages at high and low N for significant traits in experiment 2.

*, **, NS — Significant at the 0.05 and 0.01 levels, and not significant at 0.05 level respectively. Values within a row followed by the same letter do not differ significantly (p < 0.05) according to the Tukey test.

Table 3. Effects of general combining ability at high (HN) and low (LN) nitrogen levels of significant traits.

Return to the MNL Volume 79 Index
Return to the index of Maize Newsletters
Return to the Maize Genome Database Page