ABSTRACT

The patterns of gene effects on agronomic and some quality traits of eight genotypes of rice (Oryza sativa L.) and their progenies were studied using Diallel analysis method 1 involving parents, F1 hybrids and their reciprocals. The eight genotypes studied were selected for their agronomic performance and protein content from thirty genotypes evaluated in a randomized complete block design and replicated three times at the Teaching and Research Farm of the Federal University of Technology, Owerri, Nigeria during the planting seasons of 2005 and 2006. The results of the Diallel analysis conducted during the 2007 planting season indicated larger magnitude of additive components of variance in attributes like number of tillers per plant, number of days to anthesis, panicle length , number of secondary branches per panicle, 1000-seed weight, grain width and grain length/grain width ratio. Non-additive variance was more important for plant height at flowering, number of seeds per secondary branch of panicle, grain length and percentage protein content. The nature and magnitude of general combining ability(GCA) and specific combining ability(SCA) effects, as well as the direction of heterosis assessed by the SCA effects, seemed to differ for different traits and various cross combinations. No parental line consistently had negative SCA effects for all the traits. Additive gene effects with high GCA values could be exploited in WITA 4 and Max for higher tillering ability, NERICA 1, Fofifa 16 and WAB 96-1-1 for earliness; WITA 4 and NERICA 1 for production of hybrids for greater number of secondary branches per panicle; CT 7127-49 and EMPASC 105 for development of hybrids with greater number of spikelets per panicle; NERICA 1, CT7127-49 and EMPASC 105 for producing hybrids with more fertile grains; CT7127-49 and NERICAI for developing progenies with more seeds per secondary branch of panicle; IR57689-73 and Fofifa 16 for 1000-seed weight and CT 7127-49 and WITA 4 for developing long and fine grains. WITA 4 and EMPASC 105 were observed to be semi-dwarfs and offer themselves as candidate genotypes for reduction of plant height to reduce stem lodging and increase yield. The phenotypic generation means of the parental lines (P1 and P2), F1 and reciprocal cross and the BC1 and BC2, evaluated during the 2009 planting season for the study on genetic effects of the characters on the breeding lines generated, indicated that the F1 generation means were higher than their mid-parent values especially for percentage protein content. The F1 and F2 generation means were not significantly different in the majority of the cases except for lowland x upland and upland x lowland hybrids for percentage fertile spikelets and numbers of spikelets per panicle. Varied genetic effects were observed on the characters for generation mean analysis on six parameter model. Additive and dominance as well as epistatic gene effects were involved in the inheritance of most characters. Digenic interactions were significant for most of the traits. Presence of significant dominance effect combined with duplicate epitasis restricted the scope for simple selection for most of the characters. Reciprocal and maternal effects were also implicated for most of the traits.

17

INTRODUCTION

Rice (Oryza sativa L.) is a staple food for billions of people in the world. Kennedy et al. (2002) reported that it is the predominant staple food for at least 33 developing countries of the world, providing 27%, 20% and 3% of their dietary energy, protein and fat supply respectively. Muller (1984) had earlier reported that one-third of the world’s population depended on rice for over half of their caloric and protein requirements then. This may have informed why rice production, in much of the world, increasingly focuses on optimizing grain yield, reducing production costs and minimizing pollution risks to the environment. In the period of 1961 to 1981, the average world rice harvest increased from

2.23 t ha-1 to 2.86t ha-1, representing a mean increase of 23%. The usefulness of rice is indeed universal, because man benefits not only from the starchy grains but it could also be utilized in the production of starch, alcoholic beverages and soft drinks. Rice consumption in West Africa has been reported to be increasing (Watanabe, 1998), thereby, compensating for the reduction of the consumption of other cereals. Global consumption of rice continues to outpace its production. This is so because of unabated population growth. The World Bank Population Project (1995) reported that the global demand of rice then was at 524 million tonnes and projected that it would increase to over 700 million tones. In Nigeria, rice production in 2005 and 2006 farming seasons forecasts was 2.7 million metric tonnes (MT) from 2.3 million metric tonnes in 2004-2005 seasons (Oryza Market Report, 2005). The current awareness of the Nigerian government for the adoption of new rice varieties to help boost rice production is highly commendable.

Rice breeding efforts over the past three decades have been concentrated on the development of high-yielding rice varieties to meet the food need of humanity. Hybrid rice offers the potentials to boost rice yield potentials. Virmani and Peng (1999) reported

18

that it has a yield advantage of 15-20% over conventional high-yielding varieties. Choice of outstanding parents with favourable alleles will no doubt offer opportunity for greater success for higher yields. Improvements in rice quality are very crucial in meeting the demands of consumers for healthy, high quality food (Kennedy et al., 2002). Koutroubas et al. (2004) identified appearance, milling quality, cooking and processing as well as nutritional quality as the most important rice grain qualities common to all users. The nutritional quality of rice is mainly determined by the protein content of the grain. Tagwireyi and Greiner (1994) reported that 70% of the total protein for human nutrition in African regions comes from cereals. Studies have shown that protein content of rice is a quantitative trait (Singh et al., 1977). Breeding efforts for increased protein content in rice had been largely unsuccessfully owing to low heritability of protein content and complexity associated with the inheritance of triploid endosperm tissues, and its protein content has been reported to be negatively correlated with grain yield and some cooking and eating

criteria (Juliano, 1990).

Brown (1969) identified concentration of useful genes in the same genotypes as the main objective for self-fertilization of crops. Griffing (1956b), however, had earlier reported that genotypes vary in their ability to transmit desirable characteristics to their off-springs. It is on this premise, therefore, that plant breeders try to combine the desirable qualities of different varieties. The Diallel-cross method introduced by Griffing (1956a,b) and later modified by Pooni et al. (1984) and Wright (1985) is used in rice principally to determine the general and specific combining ability of the quantitative properties (Chan et al., 1990; Hoang and Tan, 1991).

Rice consumption in West Africa has been on the increase compensating for reduction in consumption of other cereals like millet, sorghum, maize etc. and thus has