+Bioline International Official Site (site up-dated regularly)

African Crop Science Journal, Vol. 9. No. 3, pp. 471-479

HETEROSIS AND COMBINING ABILITY IN A DIALLEL AMONG EIGHT ELITE MAIZE POPULATIONS

M. NIGUSSIE and H. ZELLEKE ¹

Nazret Research Centre, P. O. Box 436, Nazareth, Ethiopia
¹Alemaya University of Agriculture, P.O. Box 138, Dire Dawa, Ethiopia

Received 3 April, 2000
Accepted 12 March, 2001

Code Number: cs01066

Abstract

Crossing maize (Zea mays L.) genotypes obtained from different sources could result in better utilisation of hybrid vigor. Such heterotic response is poorly exploited in the drought-stressed areas of Ethiopia, due to problems in adaptation of some of introduced materials. This study was conducted to determine the heterosis and combining ability of eight elite maize genotypes. The eight parents were selected based on per se and top-cross performance. The parents were crossed in diallel fashion. The resulting crosses and their parents were evaluated in a randomised complete block design with three replicates at three locations for two years in Ethiopia. The combined analysis of variance showed that the mean square due to genotypes and general combining ability (GCA) were significant (P=0.01) for all the six traits studied. However, the mean square due to specific combining ability (SCA) was significant for days to tasselling, days to silking, plant height, and grain yield. The magnitude of GCA was higher than the SCA in all the cases indicating that additive gene action was more important than non-additive in the inheritance of these traits. Mid-parent heterosis occurred in varying degrees for the different traits. It was in the range of - 11.6 to 21.9% for grain yield. DTP-2 C4 and Melkasa 92 DTP1 had significant and positive GCA for days to tasselling, days to silking, plant height, and grain yield. Hence, these parents can be used to develop intermediate maturing varieties while AW-8047 significantly reduced (had negative GCA for) tasselling, silking, and plant height without affecting grain yield indicating that AW-8047 can be used as source population to develop early varieties.

Key Words: Ethiopia, crosses, heterotic, genotypes, Zea mays

Résumé

Le croisement des génotypoes de maïs obtenus des differentes sources pourraient aboutir à une meilleures utilisation de la vigeur des hybrides. La réponse hétérotique est pauvrement exploitée dans les régions à stress de séchéresse en Ethiopie, suite aux problèmes d'adaptation de quelques materiels introduits. Cette étude a été conduite pour déterminer l'hétérosis et l'efficacité de combinaison de huit génotypes élites de maïs. Les huit parents ont été séléctionnés sur base des caractères intrénsiques et la performance supérieure des croisés . Les parents ont été croisés selon le modèle diallèle. Les croisés et leurs parents ont été évalués dans un dispositif des blocs complètement randomisés avec trois répétitions dans trois localités pendant deux ans en Ethiopie. L'analyse combinée de la variance a montré que le carré moyen des génotypes et l' efficacité générale à la combinaison (GCA) étaient signifacatifs (P<0.01) pour tous les six caractères étudiés. Cependant le carré moyen de l'efficacité spécifique à la combianaison (SCA) était significatif pour les jours au tasselling, les jours au silking, la hauteur de la plante et au rendement grain. L'amplitude de GCA était supérieure plus que SCA pour tous les cas suggérant que l'action des gènes additifs était plus importante plus que les non-additifs dans l'héritage de ces caractères. L'effect hétérosis moyen du parent est apparu avec differents degrés pour ces caractères. Il variait de - 11.6 à 21.9% pour le rendement grain. DTP-2 C4 et Melkasa 92 DTP1 avaient une GCA singificative et positive pour les jours à tasselling, les jours a silking, la hauteur de la plante et le rendement grain. Ainsi ces parents peuvent être utulisés pour développer des variétés à maturité intermediaire alors que AW-8047 a réduit significativemet (avait une GCA negative pour) tasseling, silking , la hauteur de la plante sans affecter le rendement grain indiquant que AW-8047 peut être utilisé comme source de population pour développer des variétés précoces.

Mots Clés: Ethiopie, croisés, hétérotique, génotypes, Zea mays

INTRODUCTION

In the drought-stressed areas of Ethiopia, maize (Zea mays L.) is one of the most important food crops cultivated. It covers about 40% of the total arable land in the country but contributes less than 20% to the total food production (Mandefro et al., 1995). Low maize yields are mainly a result of use of local poor yielding variety due to unavailability of suitable maize varieties specifically high yielding hybrids. Fortunately, most farmers have shown an increased interest in hybrid varieties that led Nazret Maize Programme (NMP) to start work on developing hybrids for drought stressed environments of the country.

Effective selection method for grain yield and other desirable traits requires information on the magnitude of useful genetic variances present in the population, in terms of combining ability and heterosis. A suitable means to achieve this goal is the use of diallel mating systems, a method whereby the progeny performance can be statistically separated into components related to general and specific combining abilities (GCA and SCA). Combining ability analysis is one of the powerful tools in identifying the better combiners which may be hybridised to exploit heterosis and to select better crosses for direct use or further breeding work.

Although such genetic studies have been made for maize grown in other areas, little effort has been made to gather information on maize improvement for drought stressed areas of Ethiopia. In 1995, a single isolated study was done and thirty exotic maize populations were crossed to the open pollinated cultivar Katumani and the resulting top crosses were evaluated across locations. Results from 30 top crosses evaluation indicated that there was high heterosis for most CIMMYT (International Maize and Wheat Improvement Center) drought tolerant populations and negligible and/or negative heterosis was recorded for African early maturing maize (Mandefro et al., 1997). Further studies are needed. The objective of this study was to determine heterosis and estimate combining ability of eight elite maize populations.

MATERIALS AND METHODS

Based on per se and top-cross performance, eight elite maize populations were chosen to form the diallel at Melkasa (Ethiopia) during the 1997 off-season (December-April). Detailed descriptions of the genotypes are presented in Table 1. Fresh seeds of the eight parents were planted in 10 rows of 5 m length for each cross combination. All possible 28 crosses were made in both directions using bulk pollen. Seeds of each cross and reciprocal (60 cobs for each cross) were bulked for use in trials. The resulting 28 crosses and the eight parents were evaluated at Melkasa, Mieso, and Zwai (locations are described in Table 2) in 1997 and 1998 main seasons (June-November). The experimental design was randomised complete block with three replicates. The plot consisted of two 5 m rows spaced at 75 cm apart. Two plants per hill were planted first and then thinned to one plant per hill with 25 cm between hills, giving a final plant density of 53,333 plants per hectare. Phosphorus (P₂O₅) at a rate of 46 kg ha^-1 was applied at the time of planting while N at a rate 41 kg ha^-1 was applied of which 18 kg ha^-1 was applied at planting, and the remaining 23 kg ha^-1 36 days after emergence. Planting was done at the onset of the rainfall (in June). Plots were weeded on the 25^th and 40^th days after emergence. Plots were harvested and shelled manually. The shelled grain yield (kg ha^-1) was calculated after adjusting the moisture content to 12.5 percent.

Data were collected for days to tasselling, days to silking, plant height, number of ears per plant, thousand seed weight and grain yield. Analysis of variance was computed first for each location separately (data not shown) and then combined across locations. Each season at a given site was considered as a different location (i.e., the three locations by two years were considered as six locations). Analysis of variance and diallel analysis was done using a computer software suitable for Griffing's Model I, method 2, which involves parents and one set of crosses (Griffing, 1956). The mathematical model for the combining ability is

X _ijkl = μ + g_i + g_j + S_ij + b_k + e_ijkl

where X _ijkl is the performance of the cross between i^th and j^th genotypes in the k^th replicate at l^th location, μ is the population mean, gi(gj) is the GCA effect, S_ijis the SCA effect, such that S_ij = S_ji,e_ijkl is the effect peculiar to the ijkl^th observation. The restrictions ági = 0, and ásji + s_ii = 0 (for each i) were imposed. The percentage of mid-parent and better-parent heterosis was computed as outlined in Falconer (1981)

RESULTS AND DISCUSSION

Analysis of variance and genotypic mean. The combined analysis of variance showed that the mean square due to genotypes and general combining ability (GCA) were highly significant (P=0.01) for days to tasselling, days to silking, plant height, number of ears per plant, 1000 seed weight and grain yield (Table 3). The mean square due to specific combining ability (SCA) was significant for days to tasselling, days to silking, plant height and grain yield (Table 3).

The mean performance of parents and their hybrids across locations are presented in Table 4. The highest grain yield was 9870 kg ha^-1 in P1 x P4 (DTP-2 C₄ x Melkasa 92 DTP-1) followed by P2 x P4 (Kalhari Early P. x Melkasa 92 DTP1) and P1 x P5 (DTP-2 C₄x Pool 16 C₂₁) with grain yield of 9756 and 9443 kg ha^-1, respectively. The lowest grain yield (6869 kg ha^-1) was recorded for P6 x P8 (SEW-2 x TEW-GS). Among the parents, the best yield was obtained from P1 (DTP-2 C₄) while the least was from P7 (AW-8047). Thousand seed weight ranged from 383 gm for P2 x P6 (Karlhari early P. x SEW-2) to 324 gm in P4 x P5 (Melkasa 92, DTP₁x Pool 16 C₂₁ x SEW-2) for the crosses, and it varied from 357 for P1 (DTP₁ -C₄) to 305 g obtained from P7 (AW-80 47) for the parents. The number of ears per plant ranged from 1.4 for P4 (Melkasa 92 DTP₁), to 1.0 for P1 x P7. Number of ears per plant, one of the criteria used to measure stress tolerance was greater than or equal to 1.0 for all genotypes may be due to the uniform and reliable rains during the execution of the experiment.

Plant height was in the range of 233 (P1 x P2) and 177 cm (P4 x P7) while the highest ear height was 117 cm (P1 x P7) and the lowest was 73 cm obtained from P5 x P6 (Pool 16 C₂₁ x TEW-GS). The maximum number of days to silking was 62.0 days obtained from P1 x P2 or P1 x P4 and the minimum was 51.3 days obtained from P6 x P8 or P5 x P8 for the crosses while it was in the range of 64.0 (P1) to 51.0 days (P8) for the parents. Number of days to tasselling followed a similar trend as that of silking.

Heterosis. The values of heterosis over the mid-parent (MP) and the high-parent (HP) are presented in Table 5. Different hybrids showed significant heterosis for the traits studied. Eleven crosses expressed negative and significant mid-parent heterosis for days to tasselling, whereas heterosis above the high- parents was obtained for four crosses for the same trait. Heterosis above the mid- and high-parent for days to tasselling were reported by Vasal et al. (1992a). Eight crosses had significant and positive heterotic values of better-parent for days to silking. Kebede Malatu (1989) and Vasal et al. (1992b) reported better-parent heterosis for days to silking. The percentage heterosis over the better-parent ranged from 0.0% to 18.9% (P1xP7) for plant height. Significant and negative mid-parent heterosis was observed in two crosses for number of ears per plant. Crosses that exhibited negative heterosis had gene combination that reduce the number of ears per plant. Three crosses showed positive and significant heterosis above the mid-parent for 1000 seed weight while other two crosses showed negative and significant heterosis for the trait. This is in agreement with earlier findings of Shewangizaw (1983) who found heterosis above the better-parent only in few crosses for 100 seed weight. Seven crosses showed significant and positive heterosis above the mid-parent for grain yield. Heterosis above the high-parent for grain yield was recorded only for P7 x P8. The expression of grain yield above the mid-parent and high-parent has been reported by several investigators (Beck et al., 1990; Vasal et al., 1992a).

In general, heterosis or hybrid vigour was more frequent in days to tasselling and days to silking than for other traits studied in this work. Heterosis above high-parent was more variable than the mid-parent values ranging from -4.4% (P3 x P6) to 14.3% (P1 x P8) for days to silking indicating that earliness was highly enhanced in P3 x P6 while lateness was favoured in P1 x P8, and the choice from the two hybrids depends on the environmental situation for which the hybrid is targeted.

Crosses, which had divergent parents with respect to growth habit, exhibited maximum heterosis relative to the better-parent. P7 x P8 that showed maximum heterosis (21.9%) for grain yield, had parents that were different in their growth habit. This observation confirms that heterosis in maize tends to increase with increase in divergence among parents with respect to geographical origin, seed size and growth characteristics. Gallais (1984) made observations that collaborate with this principle and he also stated that extreme diversity or very related parents decreased heterosis.

Combining ability. Mean squares due to general combining ability (GCA) were significant for all traits measured (Table 3), whereas the mean squares due to specific combining ability (SCA) were significant for days to tasselling, days to silking, plant height, and grain yield implying that these traits are governed by both additive and non additive gene action. For plant height, both additive and non-additive gene actions were equally important in this study. Similar results were reported by Zambezi et al. (1986).

Several studies involving the inheritance of various quantitative traits in maize have revealed the predominance of additive gene action (Shewangizaw, 1983; Stangland et al., 1983; Zambezi et al., 1986). This has breeding implications, since additive genetic variation can effectively be exploited by simple recurrent selection. The performance of parents could be used to predict the performance of crosses. Thus, parents with good general combining ability (GCA) and per se performance can be crossed to develop high yielding composites that can be used directly for recommendation or for further breeding work.

Estimates of GCA effects of parents for various traits are presented in Table 6. Kalhari early pearl, DTP-2 C₄ and Melkasa 92 DTP₁ showed positive and significant values of GCA for days to tasselling, silking and plant height. The highest GCA value for all the three traits was obtained from DTP₂ C₄. All the remaining parents exhibited negative and significant GCA effects for days to tasselling, and silking. Three parents had negative and significant GCA values for plant height for example, AW-8047. DTP-2 C₄, and Kalhari early pearl, were good combiners for number of ears per plant while Pool 16 C₂₁ was poor combiner for number of ears per plant.

Good combiners for 1000 seed weight were DTP₂-C₄ and Kalhari early pearl. Variety AW-8047 showed significant and negative GCA value for 1000 seed weight. Positive and significant GCA effects were exhibited by DTP-2 C₄ and Melkasa 92 DTP1 for grain yield. These two parents were also good combiners for most of the traits studied. Negative and significant GCA effect were manifested by Pool 18 seq C_3,Pool 16 C₂₁, SEW-2 and TEW-GS for grain yield.

In this study, DTP-2 C4 was a good combiner for all traits. On the other hand, Pool 18 seq C_3, Pool 16 C₂₁, SEW-2, AW-8047, and TEW -GS significantly decreased days to tasselling and silking. The latter parents also decreased plant height significantly. Pool 18 seq C₃ significantly reduced grain yield. Pool 16 C₂₁ reduced significantly the number of ears per plant and grain yield while AW-8047 significantly reduced 1000 seed weight. The present results agree with the work of Hallauer and Miranda (1988).

In general, the parent DTP-2 C₄, Kalhari early pearl and Melkasa 92 DTP₁, contributed to lateness in tasselling and silking, and to increasing plant height and grain yield (except Kalhari early pearl). It is common for maturity and plant height to be associated with grain yield (Hallauer and Miranda, 1988). Pool 18 seq C3, SEW-2, AW-8047, Pool 16 C₂₁ and TEW-GS contributed earliness in days to tasselling and silking. Pool 18 seq C₃, Pool 16 C_21, SEW-2 and TEW-GS reduced grain yields significantly for the reason that earliness has inverse relation with grain yield (Bolanos and Edmeades, 1993). However, AW-8047 significantly reduced days to flowering and maturity but did not affect grain yields. This parent may have a specific trait that stimulates the physiological activities in the plant system that contribute to higher grain yields (high efficiency in conversion rate of assimilate per unit time).

Parents that are relatively late in maturity and possess drought tolerance characteristics can be used to develop varieties/hybrids that are intermediate in maturity and useful for April planting in Ethiopia. Early maturing parents are meant to develop early varieties for June planting in the drought stressed areas of the country (where the growing season is so short).

P4 x P8 was the only cross that had significant and positive SCA (data not shown) effect for days to tasselling, implying that lateness in tasselling was enhanced in this particular cross. On the contrary, P1 x P3, P2 x P3, P3 x P4, and P5 x P8 exhibited negative and significant SCA values, indicating that these crosses tasselled earlier than what would have been predicted based on their parental performance. Crosses P2 x P7, P4 x P6 and P4 x P8 had significant and positive SCA effects for days to silking showing that parents involved in these crosses enhanced lateness in female flowering. Other crosses, such as P1 x P3, P3 x P4, P3 x P6, P5 x P7 and P5 x P8, exhibited negative and significant SCA indicating that these crosses had shorter number of days in female flowering than the expected mean performance of their parents. In general, P3 and P5 enhanced earliness in male and female flowering when crossed to any other parents used in this study and may be used as good sources for earliness. Crosses P1 x P7 and P2 x P7 exhibited significant positive SCA effects for plant height, while P4 x P7 showed significant and negative SCA effect.

The two crosses that showed significant and positive SCA values for grain yield were P1 x P5 and P6 x P7, indicating that these two crosses combined well to give higher grain yields than the mean performance of their respective parents. Manifestation of good SCA may not be expected, since the parents involved in producing the crosses were broad-based populations. Several authors reported that good GCA effect is obtained from broad based parents (pools, populations, half sibs, fulsibs, composites), while good SCA is exhibited by narrow based (inbred lines) parents upon crossing.

CONCLUSIONS

The results of the study have demonstrated the importance of diallel analysis in detecting heterosis and identifying parents with general and specific combining abilities that help develop hybrids suitable for desirable traits. Additive gene action predominated in all characters studied. The dominance of additive gene action in the inheritance of quantitative traits can be effectively exploited by simple recurrent selection. Kalhari early pearl, DTP-2 C₄ and Melkasa 92 DTP1 showed positive and significant values of GCA for days to tasselling, and silking. These parents can therefore be used to develop intermediate maturing varieties to be planted in areas receiving bi-modal rainfall pattern (for April planting). Variety AW-8047 significantly reduced days to tasselling, silking and plant height without affecting the grain yields and hence AW-8047can be used in developing early maturing varieties.

ACKNOWLEDGEMENT

We are deeply indebted to National Maize Programme staff who assisted in the evaluation of these materials. We also thank African Crop Science Journal Reviewers for their comments and corrections.

REFERENCES

Beck, D.L., Vasal,S.K., Crossa, J. 1990. Heterosis and combining ability of CIMMYT's tropical early and intermediate maturity maize germplasm. Maydica 35:279-285.
Bolanos, J. and Edmeades, G.O. 1993. Eight cycles of selection for drought tolerance in low land maize II. Response in reproductive behaviour. Field Crops Research 31:253-268.
Falconer, D.S. 1981. Introduction to quantitative genetics. Longman, New York.
Gallais, A. 1984. An analysis of heterosis Vs inbreeding effects with an autotetroploid cross fertilized plant Medicogo Sativa L. Genetics 106:123-137.
Griffing, B. 1956. Concept of general and specific combining ability in relation to diallel crossing systems. Australian Journal of Biological Science 9:463-493.
Hallauer, A.R. and Miranda, J.B. 1988. Quantitative genetics in maize breeding. 2nd (Ed.). Iowa State University Press, Ames. U.S.A.
Kebede Mulatu. 1989. Manifestation of heterosis and the nature of inheritance of traits among reciprocally crossed inbred lines and population of maize (Zea mays L.). M.Sc. Thesis, Alemaya University of Agriculture. Ethiopia.
Mandefro Nigussie, Gezahegne Bogale and Benti Tolessa. 1997. Top cross performance and heterotic patterns of 31 maize populations in the drought stressed areas of Ethiopia. In: Proceedings of the Crop Science Society of Ethiopia Sebil 8:26-27.
Mandefro Nigussie, Gezahegne Bogale and Gelana Sebokesa.1995. Maize breeding for
drought stressed areas of Ethiopia. In: Proceeding of the 25th Anniversary of Nazret Agricultural Research Center: 25 years of experience in low land crops research, 20-23 September 1995. Habtu Asefa (Ed.), pp. 68-74. Nazareth, Ethiopia.
Shewangizaw Abebe. 1983. Heterosis and combining ability in 7x7 diallel cross of selected inbred lines of maize (Zea mays L.). M.Sc. Thesis, Addis Ababa University, Ethiopia.
Stangland, G. T., Russell, W.A. and Smith, O.A. 1983. Evaluation of the performance and combining ability of selected lines derived from improved maize populations. Crop Science 18:224-226.
Vasal, S.K., Srinivasan, F., Gonzalez, C., Han, G.C. and Shivaji Pandey 1992a. Heterosis and combining ability of CIMMYT's Tropical x subtropical maize germplasm. Crop Science 32:1483-1489.
Vasal. S.K., Srinivasan, G., Beck, D.L., Crossa, J, Pandey,S., De Leon, C.1992b. Heterosis and combining ability of CIMMYT's tropical late white maize germplasm. Maydica 37:217-223.
Zambezi, B.T., Harner, E.S. and Martin, F.G. 1986. Inbred lines as testers for general combining ability in maize. Crop Science 26:908-910.

The following images related to this document are available:

Photo images