Multivariate analysis of grain amaranth (Amaranthus spp.) accessions to quantify phenotypic diversity

Abstract

Twelve amaranth accessions were evaluated in a randomized complete block design with three replications in the research field of the Institute of Agriculture and Animal Science, Lamjung, Sundarbazar, Lamjung, Nepal from March to July 2021 to assess the phenotypic diversity. Principal Component Analysis (PCA) showed that only two principal components were significant with their eigenvalues >1 and combinedly accounted for 88.3% of the total variance. PC1, which explained 71.9% of the variance, was highly and positively contributed by days to 50% inflorescence, plant height, leaf length, leaf width, petiole length, 1000 seed weight, and grain yield. PC1 was negatively affected by the number of primary branches per plant and the number of leaves per plant. PC2, which explained 16.4% of the variance, distinguished plants with high number of leaves and a higher inflorescence length. The accessions were grouped into 4 clusters. Cluster 2 had the greatest intracluster distance (D2 = 13.19), while Cluster 3 and 4 had the greatest intercluster distance (D2 = 21.73), followed by Cluster 2 and 3 (D2 = 17.37). Cluster 1 had the highest number of leaves per plant and the lowest yield. Cluster 2 had the maximum grain yield and plant height. Cluster 3 had the lowest inflorescence length, and cluster 4 had the highest leaf length and 1000 seed weight.

Keywords

• Amaranthus
• Cluster analysis
• Diversity
• Mahalanobis distance
• PCA