DECODING GENETIC VARIABILITY, TRAIT ASSOCIATIONS AND MULTIVARIATE DIVERGENCE TO IDENTIFY ELITE PARENTS IN CABBAGE (BRASSICA OLERACEA VAR. CAPITATA L.)

Authors

  • Abinash Mishra Author
  • Anita Mohanty Author
  • Pradyumna Tripathy Author
  • Dibya Sundar Kar Author
  • Ankita Mishra Author
  • Disha Baishnabi Author
  • Akash Kumar Parida Author
  • Aurobinda Behera Author
  • Jyoti Ranjan Das Author
  • Rohitash Kumar Author
  • Dibyajita Swain Author

DOI:

https://doi.org/10.4238/hrpc0s04

Keywords:

Brassica oleracea var. capitata; Heritability; Path coefficient; Mahalanobis D²; Principal component analysis; Parental selection

Abstract

Choosing parents that are both high yielding and genetically distant is the starting point of any cabbage hybrid programme. Eighteen cabbage genotypes (thirteen advanced AICRP lines and five cultivars) were evaluated for fifteen quantitative traits in a randomized block design with three replications during the rabi season of 2022–23 at Bhubaneswar, India. The genotypes differed significantly (P ≤ 0.01) for fourteen traits; plant height was the exception. The top yielders were 2022/CABHYB-1 (295.33 q/ha), Pragati (284.58 q/ha) and 2021/CABHYB-5 (278.88 q/ha), which were statistically on par, followed by 2021/CABHYB-1 (252.78 q/ha). Net head weight showed the highest genotypic and phenotypic coefficients of variation (30.48 and 31.55%). Net, marketable and gross head weight and yield had high heritability (91.0 93.3%) together with high genetic advance as per cent of mean (47.1–60.6%), which points to mainly additive gene action. Yield was positively and significantly correlated with frame spread east–west (r = 0.64), gross head weight (0.63), marketable head weight (0.58), equatorial diameter (0.54), polar diameter (0.52), net head weight (0.49) and frame spread north–south (0.48) and negatively with stalk length (–0.49). In the path analysis, frame spread east–west (0.522), polar diameter (0.298) and marketable head weight (0.222) had the largest positive direct effects on yield (residual effect 0.548). Mahalanobis D² analysis placed the genotypes in six clusters and clusters I and VI were the farthest apart (D² = 1377.16). Five principal components with eigenvalues above one explained 88.72% of the variation. The first component (39.73%) was a head weight–yield axis on which 2022/CABHYB-1 and Pragati scored highest. Ward clustering agreed moderately with the Tocher grouping (adjusted Rand index = 0.38). Frame spread and head weight can therefore be used as selection criteria and 2022/CABHYB-1, Pragati, 2021/CABHYB-5 and 2021/CABHYB-1 are suitable parents for crossing with genotypes of the divergent clusters I and III.

Downloads

Published

2026-11-04

Issue

Section

Articles