GENETIC DISTANCE, HERITABILITY AND CORRELATION ANALYSIS FOR YIELD AND FIBRE QUALITY TRAITS IN UPLAND COTTON GENOTYPES

Abstract

The present research was carried out to determine the genetic distance, heritability and correlation coefficients for yield and fibre quality traits in fifteen upland cotton genotypes. The traits studied were; plant height, sympodial branches plant-1, number of bolls plant-1, boll weight, ginning outturn (%), seed index, seed cotton yield plant-1 and staple length. The genetic distance was further supported by statistical procedures like cluster analysis, coefficient of parentage and principal component analysis. The mean squares from analysis of variance revealed that genotypes differed significantly at 5 % probability level for all the traits except that seed index and staple length were non-significant. The mean performance of genotypes suggested that IR-524 produced the tallest plants; Sadori produced the maximum number of sympodial branches plant-1, Sohni formed highest number of bolls plant-1; Tarzan-1 yielded heaviest bolls and gave highest seed cotton yield plant-1; NIA-Ufaq ginned maximum lint % and Chandi-95 recorded highest seed index and measured longest staple length. The correlation coefficient indicated that all the morpho-yield traits were significantly associated with seed cotton yield and with each other suggesting that one or more yield traits can be used as selection criteria to improve the seed cotton yield.
The higher heritability estimates (h2b.s) were recorded for almost all the studied traits except seed index. Such results revealed that improvement in those traits can be made through direct phenotypic selection. Cluster analysis arranged all the 15 genotypes into four different groups. Cluster third characterized the genotypes as having more number of plants and with higher seed cotton yield plant-1, hence genotypes of this group may prove promising in producing higher seed cotton yield plant-1. While concerning genetic distance, out of 105 pair wise comparisons, some pairs manifested greater genetic distance; consequently, those genotypes can be utilized in heterosis and backcross breeding programmes for the introgression of novel genes. From principal component analysis (PCA), eight components were extracted for studied characters. The first two components accounted for about 99 % variation from total variation, thus the cumulative percent of variance was 99.0 % in first two components. The first two PCAs demonstrated significantly higher variability that can be exploited for further breeding programmes in cotton.