Heterosis and Combining Ability for Grain Yield and Yield Component Traits of Maize in Eastern Ethiopia

Currently, concentrated efforts are underway in the country by different maize research institutions to move more toward the use of hybrids varieties. To fulfill this aim, combining ability studies have prime importance in maize hybrid development since it provides information for the selection of parents and also provides information on the nature and magnitude gene actions. The two types of combining ability, general (GCA) and specific (SCA) have been recognized in genetic studies. General combining ability relates to additive gene effects, while specific combining ability reflects the non-additive gene actions (Sprague and Tatum, 1942). GCA is average performance of a parent in a series of crosses and SCA designates those cases in which certain combinations perform relatively better or worse than would be expected on the basis of average performance of parents. The GCA includes additive and additive × additive variances, while SCA is responsible for non-additive genetic variances (Ibni Amin Khalil, 2010). In specific combining ability; dominance or epistatic effects of genes are commonly involved in maize (Rahman, 2013). With this aim in mind, this research was conducted with the objective to study heterosis and combining ability in maize (Zea mays L.) for yield and yield related traits in eastern Ethiopia.

Materials and Methods

Planting Materials

The experimental materials comprised of 28 F1 hybrids obtained from a cross between 8 inbred lines in a diallel fashion plus the parental lines and one standard check (BH 660). The description of the lines is depicted in table 1. All the lines have been developed by Haramaya University Maize Research Program, Ethiopia. F1 hybrids have been obtained by crossing in 8 x 8 diallel mating design in the main cropping season of 2010.

Methods

The 28 F1 hybrids plus the parental lines were tested in 2011/2012 and 2012/13 cropping seasons for two years at Haramaya University main campus (Raare research site) using randomized complete block design with three replication. Each plot consisted of four rows with row length of 5.1m. Tow Seeds were sown per hill and at 4-5 leaf stage thinned to one plant per hill. The distance between rows and between plants with in the row was maintained at 0.75m and 0.30m, respectively. At sowing, P2O5 at the rate of 46kg/ha in a form of DAP and 35 kg/ha of N in a form of urea was applied. At about booting stage, 35 kg/ha N was also applied. All other field management practices including stalk borer and armyworm control were done as required. Data were collected from the middle two rows leaving the end hills. Data were collected on days to maturity, ear length, ear diameter, number of rows per ear, number of kernel rows per ear, 1000 kernel weight (g) and grain yield in t/ha. Before analysis of the data, both grain yield and 1000 kernel weight were adjusted to 12.5% moisture content.

Statistical Analysis

All data were subjected to simple analysis of variance to see the existence of genetic variability whether there are differences between the tested genotypes before running combining ability analysis.

Analysis of variance for general combining ability (GCA) and specific combining ability (SCA)

was done following Method II, Model I as suggested by Griffing (1956).

Heterosis expressed as increase or decrease of F1  hybrid value over mid-parent (relative heterosis), better parent (heterobeltiosis) and over the best commercial check (standard heterosis) were calculated for each character using the following formulas suggested by Hayes et al. (1955).

Where: F1=mean performance of F1, MP = mean mid-parental value = (P1 + P2)/2, P1= mean performance of parent one, P2 = mean performance of parent two, BP= mean performance of better parent, CC= mean performance of the best commercial check.

Results and Discusions

Effect of Genotypes on the Genetic Variation of Maize

The analysis of variance showed highly significant differences between the genotypes for grain yield, days to maturity and yield component traits indicating there is sufficient genetic variation among the tested genotypes. Several maize researchers also found statistically significant differences between the genotypes for grain yield and yield related traits in their study on heterosis and combining ability of maize (Dagne et al., 2007; Amiruzzaman et al., 2010;

Table 1: List of Lines used in the Study.

Source: Haramaya University Maize Research Program

Amanullah et al., 2011 and Shushay et al., 2014). Days to maturity and grain yield showed statistically highly significant difference between the genotypes in their response to seasonal variation/year of testing. Significant interaction between genotype and year of testing was detected for 1000 kernel weight and grain yield indicating the need for testing varieties/genotypes over locations and seasons in order to see their performance under the different environments (Table 2). Similar result was also noticed by Dagne et al., 2007).

Table 2: Mean Squares due to Grain Yield and Yield Components of Maize (Zea mays L.) in Eastern Ethiopia.

*, ** = significant at 0.05 and 0.01 level of significant, respectively and d.f. = degrees of freedom, DM = Days to maturity; EL = Ear length; ED = Ear diameter; KRE = Kernel rows per ear; KR = Kernels per row; TKWT = Thousand kernel weight; GY = Grain yield

Mean Performance of Genotypes F1 Hybrids and Parents

There were significant differences between the genotypes for grain yield (table 3). The yield of the crosses ranged from 7.15 – 10.15 kg/ha. Top yielding crosses were L1 x L4 (10.15 t/ha), L1 x L5 (10.05t/ha), L7 x L8 (10.04 t/ha), L3 x L8 (9.60t/h), L3 x L5 (9.55t/ha), L1 x L7 (9.48 t/ha), L3 x L6 (9.16 t/ha, L3 x L7 (9.01 t/ha) produced over 9 t/ha of grain per hectare, whereas the yield of lines ranged from 4.53-6.12 t/ha (Table 3).

Table 3: Mean performance of maize crosses and lines for grain yield and yield components of maize (Zea mays L.) in eastern Ethiopia.

Table 3(B): Mean performance crosses and lines …

Statistically significant differences (LSD at P0.01 level of significance) have been observed for all yield related traits (Table 3). Days to maturity ranged from 157-173. Almost all the hybrids were relatively earlier that the parents. L8 was the latest to reach maturity, where as the hybrid L2 x L3 was the fastest to mature. Ear length ranged from 14.08-18.08cm. Inbred lines exhibited shorter ear length compared to the hybrids. Ear diameter ranged from 3.38-4.49 cm; average kernel rows per ear ranged from 10 – 12; kernels per row from 27 – 41 and 1000 kernel weight from 138 – 408 grams obtained by L1 and L3 x L5, respectively. The top yielding hybrids also showed medium to high values for yield components.