To address production problems such as low seed setting rate， in this study， a total of 300 Leymus chinensis accessions were evaluated across three consecutive years of field trials and systematically measured their four growth traits： Plant Height of Reproductive Branch at Heading Stage （PH_RBHS）， Plant Height of Reproductive Branch at Maturity Stage （PH_RBMS）， Leaf Length （LL）， Leaf Width （LW）， and eight seed setting-related traits： Spike Weight （SW）， Spike Length （SL）， Number of Spike Nodes （NSN）， Seed Setting Rate （SSR）， Number of Spikelets （NS）， Floret Number per Spikelet （FNPS）， and Seed Number per Spike （SNPS）. Diversity analysis and cluster analysis jointly revealed a rich genetic background within the population： the coefficient of variation ranged from 10.73% to 51.86%， and the genetic diversity index ranged from 1.853 to 2.068； the 300 accessions could be classified into four distinct clusters. Correlation analysis indicated that seed setting rate （SSR） was correlated to varying degrees with 11 traits， such as Spike Weight （SW）， Spike Length （SL）， and Seed Number per Spike （SNPS）. The first four principal components accounted for 79.26% of the total variance， effectively capturing most of the information of the original traits. Finally， based on the best linear unbiased estimate （BLUE） of phenotypic data， a comprehensive evaluation was conducted using a membership function analysis that integrated heading rate （HR） and seed setting rate （SSR）， and 50 elite accessions were identified. This study provides support for further gene mining， genotype-phenotype association analysis， and molecular breeding.

Leymus chinensis  /  Phenotype  /  Seed setting rate  /  Comprehensive evaluation  /  Elite germplasm accessions