Morphological characterization of newly introduced lettuce (Lactuca sativa L.) germplasm through principal component and regression analyses

Abstract

Twenty seven newly introduced genotypes of lettuce collected from different indigenous and exotic sources were characterized with respect to economically important traits by using principle component and regression analyses. Significant differences were observed among the genotypes for all the characters under study. CGN-10944 performed best among all the genotypes for most of the traits under study. Principal component analysis indicated that the first five factors had variances of 3.88, 2.78, 1.47, 1.42 and 1.24 with 25.87%, 18.57%, 9.81%, 9.49% and 8.26% of total variation, respectively and aggregating to 72.01% of total variation. The highest positive values of different traits in different components indicated that main emphasis should be given on the gross head weight (X 2 ), polar (X 6 ) & equatorial diameter (X 7 ) and seed vigour index-I (X 12 ) & II (X 13 ) to have higher yields in lettuce. To quantify the importance of each variable in predicting yield, multiple linear regression model was developed. The coefficient of determination (R 2 ) in linear regression model revealed that 94% of total variation in yield per plot (X 15 ) was influenced by days to marketable maturity (X 1 ), gross head weight (X 2 ), heading percentage (X 4 ), seed germination (X 11 ), seed vigour index-II (X 13 ) and severity of bacterial soft rot (X 14 ). Therefore, main emphasis should be given on days to marketable maturity, gross head weight, polar & equatorial diameter, heading percentage, seed germination, seed vigour index-I & II and severity of bacterial soft rot during selection and hybridization programmes for yield improvement in lettuce.