Articles
MULTIVARIATE ANALYSIS OF MAJOR QUANTITATIVE TRAITS OF AN ORIENTAL LILY SEEDLING POPULATION
Article number
953_39
Pages
285 – 291
Language
English
Abstract
In order to learn more about the relationship between major quantitative traits, and obtain uncorrelated new variables of reduced dimensions for further comprehensive evaluation of different genotypes, we conducted a multivariate analysis of major quantitative traits of the Oriental lily seedling population.
By means of correlation analysis and principal component analysis, we studied 15 quantitative traits (6 vegetative traits and 9 reproductive traits) of 90 F1 hybrids of Constanta × Sorbonne. The vegetative traits observed included plant height, basal diameter, leaf density, mid-leaf length, mid-leaf width, and pedicel length; and the reproductive traits are crown diameter, inner petal length, inner petal width, outer petal length, outer petal width, ovary length, style length, anther length, and filament length.
Among all the investigated traits, plant height had the greatest variation.
Correlation analysis showed a moderate correlation between the reproductive traits and vegetative traits.
Neither plant height nor basal diameter had a significant relationship with any reproductive traits.
Among the vegetative traits, either plant height or basal diameter correlated with leaf traits moderately; and among the reproductive traits, there was a significant relationship between the inner petal length and outer petal length, outer petal width and inner petal width, ovary length and style length.
Principal component analysis revealed that the first four principal components with eigenvalues >1 explained about 71.44% of the total variance.
The first principal component represents the crown size variable, and the second principal component is the stem variable, each having a respective contributory ratio of 36.39 and 19.08%.The third and fourth principal component can be characterized as the pistil and leaf density variable, respectively.
By means of correlation analysis and principal component analysis, we studied 15 quantitative traits (6 vegetative traits and 9 reproductive traits) of 90 F1 hybrids of Constanta × Sorbonne. The vegetative traits observed included plant height, basal diameter, leaf density, mid-leaf length, mid-leaf width, and pedicel length; and the reproductive traits are crown diameter, inner petal length, inner petal width, outer petal length, outer petal width, ovary length, style length, anther length, and filament length.
Among all the investigated traits, plant height had the greatest variation.
Correlation analysis showed a moderate correlation between the reproductive traits and vegetative traits.
Neither plant height nor basal diameter had a significant relationship with any reproductive traits.
Among the vegetative traits, either plant height or basal diameter correlated with leaf traits moderately; and among the reproductive traits, there was a significant relationship between the inner petal length and outer petal length, outer petal width and inner petal width, ovary length and style length.
Principal component analysis revealed that the first four principal components with eigenvalues >1 explained about 71.44% of the total variance.
The first principal component represents the crown size variable, and the second principal component is the stem variable, each having a respective contributory ratio of 36.39 and 19.08%.The third and fourth principal component can be characterized as the pistil and leaf density variable, respectively.
Publication
Authors
F. Qi, L. Fang, M. Xi, S. Qiu, J. Shi
Keywords
F1 hybrids, quantitative traits, correlation analysis, principal component analysis
Online Articles (49)