Title : Genetic architecture of yield and its components in a spring x winter diallel cross of wheat (Triticum aestivum L.) Name of the student : Peer Arif Sultan Guide : Shafiq A. Wani Degree : PhD Year : 2004 Key words : Combining ability, genetic components of variance, graphical analysis, variability, correlation, path analysis, Spring and Winter wheats (Triticum aestivum) ABSTRACT Genetic architecture of yield and its attributing traits were studied in a full spring x winter diallel cross of wheat (Triticum aestivum L.) using five winter and five summer ecotypes. The experimental materials, evaluated over two random environments in a completely randomized block design revealed considerable variability for days to 50% heading, plant height, productive tillers plant-1, leaf area index, spike length, spikelets spike-1, grains spike-1, 100-seed weight, days to maturity, harvest index and grain yield plant-1. Significant G x E interaction was also observed for most of the traits including grain yield plant-1 revealing that the behaviour of the parents and crosses was not similar in the set of random environments. Different sets of gene complexes were observed to affect the yield component and maturity traits. Phenotypic and genotypic coefficient of variation was high for productive tillers plant-1; moderate for plant height, spikelets spike-1, grains spike-1, 100-seed weight and grain yield plant-1; and low for days to 50% heading, days to maturity, leaf area index, spike length and harvest index. Broad sense heritability was high for days to 50% heading, plant height, spikelets spike-1 and 100-seed weight ; moderate for productive tillers plant-1, spike length and harvest index, whereas, narrow sense heritability was high only for plant height and moderate for productive tillers plant-1, leaf area index and 100 seed weight. The low additive genetic heritability (breeding value) is expected in the F1 hybrids because of high level of heterzygosity at loci arising from the coming together of the different gene complexes. On the basis of data pooled over environments, genetic gain was low for almost all the characters indicating the role of environment in the expression of characters and higher magnitude of heterotic behaviour in the crosses. This genetic gain is expected to increase in the subsequent generations after following selective mating system (bi-parental mating). Grain yield exhibited significant positive genotypic and phenotypic correlation with all the component traits studied. However, it was negative and significant with maturity traits at both the levels. This significant negative correlation of grain yield plant- 1 with maturity traits might have arisen since under Kashmir valley conditions the temperatures abruptly increase during May/June, which enhances the maturity (flowering & physiological maturity). This needs to be further investigated into before, scientific conclusions could be drawn for selection of early maturing high yielding genotypes. For yield components, significant positive association was revealed between the traits and similar inter-relationship was observed with leaf area and harvest indices. Based on the study of cause and effect relationship, the direct contribution to grain yield was observed from grains spike-1, productive tillers plant-1, plant height, and leaf area index. High positive indirect effects were, however, observed via grains spike-1 and productive tillers plant-1 in almost all the traits. The improvement in grain yield through selection for yield component traits via., grains spike-1 and productive tillers plant-1 would result in the isolation of putative high yielding genotypes in the present set of materials. None of the parents exhibited significant and desirable gca effect for all the traits. The parents with desirable gca effect for grain yield plant-1 were SKW-192, SKW-209, SKW-210 and SKW-211. For maturity traits the spring wheat parents demonstrated earliness. Similarly, among the cross combinations none revealed the desirable sca performance for all the traits. However, several cross combinations were identified to be superior for various yield and maturity traits based on sca and per se performance. Five top ranking and desirable crosses on the basis of sca performance for grain yield were observed to be HS-240 x SKW-210, HS-240 x SKW-209, SKW-210 x HS-295, SKW- 209 x HS-365 and SKW-211 x HS-240. These crosses were mostly the result of high x average or average x average general combiners. On the basis of per se performance five top ranking cross combinations for grain yield included HS-240 x SKW-193, HS-295 x SKW-192, SKW-211 x HS-295, VL-845 x SKW-210 and HWP-220 x SKW-211. By and large, there was no correspondence between sca and mean per se performance for almost all the traits. Manifestation of desirable sca effects due to contribution of more favorable alleles and their interactions, could be used to generate desirable segregants. The two best crosses possessing desirable sca effect for grain yield plant-1 and productive tiller plant-1 were HS-240 x SKW-193, HS-295 x SKW-192 and HS-240 x SKW-193 and SKW-209 x HS-240. Graphical and component analysis of genetic variance revealed the significance of additive gene action for plant height, productive tillers plant-1, leaf area index, days to maturity and grain yield, whereas, components of variance due to dominance deviation were significant for all the traits indicating the preponderance of non-additive gene action in the inheritance of various traits. Average degree of dominance was in the over dominance range for all the characters but in the graphical analysis their dominance for most of the traits was in partial to complete dominance range. This upward bias in the dominance variation in the component analysis was observed to have arisen due to epitasis and/or linkage effects. Dominance was unidirectional for grain yield, spikelets spike-1, grains spike-1 and maturity traits, whereas the gene distribution was observed to be asymmetrical for all the traits except grain yield and harvest index. However, epistasis appeared to govern the inheritance of days to heading, spike length, spikelets spike -1 and grains spike-1. Crossing of the spring x winter wheat genotypes would broaden the genetic base of the population for selection of potent homozygous lines possessing high yield potential. Further, introduction of elite winter wheat genotypes and their exploitation through winter x spring wheat crosses is suggested to generate larger spectrum of variability for use in developing suitable genotypes for high altitude Valley conditions by following a selective diallel mating system to slow the process of the fixation of alleles.