Selection of consistent plant traits for sunflower growth using principal component analysis

Abstract

Principal component analysis has been explored for assessing the consistency of 14 plant traits for sunflower growth using multivariate data of 6 field experiments conducted during kharif 1994 to 1999 seasons on rainfed alfisol. Seven genotypes have been used in the study viz., Guj-sun-1, MSFH-8, MSFH-17 KBSH-1, Jwala, Pac-36 and Morden. Observations were recorded on leaf nitrogen, leaf area, leaf weight, leaf number, stem nitrogen, stem weight root length, root weight, stomatal conductance, photosynthesis, total biomass on 30, 45 and 60 days after sowing (DAS) and flower head diameter and flower head weight on 45 and 60 days after sowing in each season. The first two principal components have extracted about 80% of variance in the data of different plant traits on different days after sowing. The consistency of plant traits has been assessed by examining correlations between different plant traits and the distribution of loadings of plant traits on the first two leading principal components 30, 45 and 60 days after sowing. The results have indicated that stem weight and lead number on the 30th day, leaf area and leaf weight on the 45th day and leaf area, leaf weight and total biomass on the 60th day after sowing had significantly higher loadings on the 1st principal component. Similarly, root length on the 30th day, flower head diameter and flower head weight on the 45th day and root length and photosynthesis on the 60th day after sowing had significantly higher loadings on the 2nd principal component. Based on a graphic plot of the loadings, root length (30 and 60 DAS), stomatal conductance (30, 45 and 60 DAS) and photosynthesis (30 DAS) were found to be consistent since their loadings on the 1st principal component had high mean values with low standard deviations. Similarly, stem nitrogen (45 and 60 DAS), stomatal conductance (45 DAS), photosynthesis (30 DAS), leaf nitrogen (60 DAS), root length (30 DAS), flower head diameter (45 DAS) and flower head weight (45 DAS) were also consistent since their loadings on the 2nd principal component had high mean values with low standard deviations.