Comparison of the gene action controlling metric characters in differing tyes of Upland cotton, Gossypium hirsutum L

Abstract

This experiment was designed to determine if the same genes or blocks of genes operate and respond in the same ways in differing plant types. Six selected breeding stocks of Upland cotton, ranging in stature from semidwarf to normal, were evaluated. The experimental design was a randomized complete block with a split plot arrangement, with two replications. An F(,2) half-diallel cross analysis was used to investigate the actions and interactions of sixteen agronomic and lint-quality characters. Estimates of the genetic parameters D, F, H(,1), H(,2), and h('2) were made. A significant additive variation was found, except for total boll number, on at least one level of the factors used in the analysis of variance for each character where significant genotypic differences were found. Significant dominance was also found for most characters, but not for node of first flower, total boll number and second harvest boll number. Genetic component ratios were estimated from the genetic parameters to determine the average degree of dominance present, the average frequency of positive to negative alleles, the ratio of dominant to recessive genes, an estimate of the number of dominant genes or groups of genes that are involved in the control of the character, and an estimate of heritability. All of the characters, except 2.5% span length, showed overdominance. The yield characters, node of first flower, elongation, and micronaire showed asymmetry of gene distribution. There was an indication of an excess of dominant to recessive genes present for all characters, except 2.5% span length and strength. In all cases only one group of genes showing dominance was estimated to be acting. Heritability estimates were low for all characters except seed index. Epistasis was found for all characters measured. It was of greatest influence in node of first flower, seed index, lint percentage, 2.5% span length, strength and elongation. The diallel analysis showed interaction of genotypes with environment for the flowering and yield characters. However, the pattern of response for each of these characters was very similar. For lint-quality characters the response was stable over environments even though epistasis was greatest among this groupe. The results of this study suggest that the same genes or blocks of genes are operating and responding in the same way within differing plant types over different environments.