Abstract
Lotus corniculatus L. is a perennial legume valued for its roles in forage production, soil and water conservation, and landscaping. However, the limited number of improved cultivars hampers its broader utilization. In this study, we characterized the genetic variation of 23 accessions from diverse geographic origins using 12 quantitative traits and 29 simple sequence repeat (SSR) markers. Quantitative trait analysis revealed substantial variation across tested traits, and significant correlations were observed between nutritional quality and morphological traits, which suggests that phenotypic selection can indirectly enhance forage quality. Cluster analysis based on phenotypic and nutritional data grouped accessions into five categories, identifying key germplasm types with distinct breeding advantages, such as high crude protein and ether extract content (Q1), superior leaf morphology (Q2), and high total sugar content with thick stems (Q5). Genome-wide SSR mining identified a large number of loci, dominated by dinucleotide repeats. The 29 highly polymorphic SSR primers used exhibited strong polymorphism, and accession-level SSR-based clustering separated the accessions into three groups that broadly aligned with geographic origin. Analysis of molecular variance (AMOVA) indicated that most genetic variation resided within populations, underscoring the potential for intra-population selection. These findings establish a germplasm grouping basis that integrates phenotypic performance, nutritional quality, and genetic background. The results offer practical pointers for identifying complementary parental combinations, while emphasizing that multi-environment, multi-year validation and targeted crossing are required before generalization or breeding deployment. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12870-025-07914-8.