Studies on quantitative inheritance in Nicotiana tabacum L. II. Components of genetic variation for flowering tome, leaf number, grade performance and leaf burn

Abstract

The components of genetic variation for two characters, flowering time and the number of curable leaves, which influence the gross yield, and two quality characters, grade performance and burning quality of the cured leaf, were examined in the F² and back-cross generations of a set of diallel crosses between four flue-cured varieties selected on the degree of divergence as measured by Mahalanobis's D² statistic. Differences in the relative proportions of D and H components were observed, in different crosses, for the same character. These differences were parallel to the degree of divergence between the parents concerned. Significant non-allelic interaction of a generalised nature was detected for each of the four characters. Additive component formed a substantial purt of the total genetic variation for the character curable leaf number only. A comparison of the degree of divergence between the parents and the heritability estimates for the characters concerned, indicated that in a majority of the cases, the use of D² statistic for the choice of the parents would be useful for evaluating the potential of a cross. Analysis of the diallel material for Vr and Wr components confirmed the existence of a high degree of non-allelic interaction. The distribution of the desirable genes, for quality characters, over several parents, suggests the need for multiple cross for the accumulation of a maximum number of favourable genes. A contrast in the nature of gene action between quality characters and gross yield components was also indicated in the study. A variety Hicks was found to be unique in its high general combining ability and high additive variance for all the four characters, in its crosses, whereas another variety Harrison Special was a consistently poor combiner. A comparison of the components of genetic variability estimated by other workers with those of the present investigation was made with reference to the limitations of conclusions from one or few crosses. The utility of the combined use of D² statistic, the analysis of Vr–Wr components and the partition of the components of genetic variation was discussed with reference to the future breeding work in flue-cured tobacco for the characters under study.