Wuling Mountainous Maize Inbred Lines Genetic Characteristics And Breeding Potential

Fourteen maize inbred lines and forty-nine crosses mated by NC II design were used in the experiment. Their genetic characteristics and potential in breeding were analyzed with mixed linear model analysis, combining ability analysis, genetic clustering analysis, synthetic heterosis of multi-character analysis of seventeen characters. The major results are as following: The results revealed by mixed model analysis and combining ability analysis are compared and analyzed. The mixed linear model analysis indicated: As for kernel weight/plant, kernels/row, 100 kernels weight, ear diameter, ear height, width of ear leaf, filling period , the genetic model of theirs is dominant-epistatic model. As for growth period duration, number of tassel branch, tassel weight, length of fruit stalk, bud number, bud length, bud weight, the genetic model of theirs is additive-dominant model. As for rows/ear, the genetic model of its is additive-dominant-epistatic model. But combining ability analysis is based on additive-dominant model. It reveals that some traits just have additive effect and dominant effect. The differences of genetic variance component between mixed model analysis and combining ability analysis are significant. It is suggested by combining ability analysis that the differences of genetic variance component in some characters are mainly caused by less estimating genotype-environment interaction, and/or by more estimating additive effect in the others. The difference of heritability in the broad sense is caused by genotype- environment interaction. The difference of heretability in the narrow sense is caused by genotype-environment interaction and/or additive effect/A-A effect estimated by the two different methods. There is no significant difference between grain weight per plant and its 34 predicted value by LUP(linear unbiased prediction) and AUP(adjusted unbiased prediction). Array sum of parents(including parents effect) comply with GCA in number order(from high to low) Correlation coefficient between the dominant effect of LUP and SCA(special combining ability) /heterosis are 0.70 and 0.93 respectively at 0.01 significant level. Correlation coefficient between SCA and heterosis is just 0.56 at significant 0.01 level. Results from genetic clustering indicate the classification of parents by their crosses mated is more capable of revealing the genetic relationship and the most efficient method is classification by GCA(general combining ability). But clustering results of maize inbred lines by their characters just show their difference of phenotype. And there is no inevitable inner relation between clustering result and consanguinity or geographical source of them, whichever methods are used.