Using 77 Cymbidium ensifolium germplasms as the test materials, the phenotypic diversity of 21 quantitative traits and 12 qualitative traits was analyzed. The results showed that the coefficient of variation (CV) for quantitative traits ranged from 14.72% to 42.79%, indicating a significant degree of variation. Among these, the CV value for flower length was the lowest (14.72%), suggesting that this trait was relatively stable across different germplasms. The CV value for the number of flower stalks was the highest (42.79%), followed by the number of flowers (41.67%). Additionally, the Shannon-Wiener index (I) was close to 5, indicating a high level of diversity in quantitative traits among the germplasms. In the binary analysis of qualitative traits, the distribution of sepal spots or stripes (colors) among the germplasms was similar, with a rich variety of expressions for the features on the flower sepals. The CV for qualitative traits ranged from 9.86% to 83.81%, also showing a significant degree of variation. Among these, the CV value for leaf posture was the lowest (9.86%), while the CV values for lip shape (83.81%) and sepal color (70.69%) were the highest. Furthermore, the I index values for qualitative traits ranged between 4 and 5, indicating a high level of diversity in qualitative traits among the germplasms. The Simpson index (D) for qualitative traits ranged from 0.985 to 0.991, close to the value of 1, suggesting a relatively uniform distribution of qualitative traits and a high degree of diversity among the germplasms. Principal component analysis was conducted on 33 phenotypic traits of the germplasms, and the first 10 principal components were extracted based on the criterion that the characteristic root value is greater than 1. The characteristic value of the first 10 principal components was 6.406, 4.475, 2.840, 2.425, 2.139, 1.619, 1.517, 1.430, 1.163, and 1.038, with a cumulative contribution rate of 75.917%. The number of flower parts among the phenotypic traits had a significant weight in the principal components, indicating it is a major phenotypic trait of C. ensifolium. Hierarchical clustering analysis was performed on the germplasms using Euclidean distance, resulting in a dendrogram. At the 15th Euclidean distance, the germplasms were clustered into five major groups, with the main features of phenotypic traits concentrated in quantitative traits, especially in the significant weight of the number of flower parts, consistent with the results of the principal component analysis. This study explored the genetic diversity and relationships among C. ensifolium varieties through morphological trait diversity analysis, providing a basis for the scientific management and conservation of C. ensifolium germplasm resources. It also offers valuable genetic materials for future C. ensifolium breeding work.

Cymbidium ensifolium  /  quantitative traits  /  qualitative traits  /  principal component analysis  /  cluster analysis