Abstract: 【Objective】Alfalfa is the most important cultivated forage across the world. However, for little improvement in forage yield and quality in recent years, conventional breeding methods are far from satisfying the practical need. On the one hand, the improvement in cultivars relies on the quantity of breeding resources, on the other hand, it relies on the understanding of the genetic basis of agronomic traits. Based on SSR markers, the study was conducted to analyze the genetic diversity and population structure of the germplasm from China and the United States and provide basic information on mining beneficial markers and alleles significantly associated with important quantitative and quality traits of alfalfa when using genome-wide association study to facilitate breeding process.【Method】In total, 40 SSR loci, covered the entire alfalfa genome with 3 to 9 SSR loci chosen on each chromosome, were used to conduct the genomic scanning of a collection of 100 Chinese and American alfalfa genotypes out of 16 main cultivars and to determine the genetic diversity and population structure. For Chinese cultivars, 8 genotypes were chosen for each cultivar; while for American cultivars, 4 genotypes per cultivar. Based on the admixture model, the software program STRUCTURE was used to determine the population structure of alfalfa. The optimal value of K (the number of clusters) was deducted by evaluating K=1-6. Length of burn-in of the Markov Chain Monte Carlo (MCMC) iterations was set to 10 000 and data were collected over 100 000 MCMC iterations in each run. Ten iterations per K were conducted. The optimal value of K was identified using two methods. The result was displayed by Structure Harvester and Distruct. Further, the principle component analysis and clustering analysis were conducted to justify the result of population structure. Based on the result of population structure, the genetic diversity of all samples and two populations were analyzed.【Result】A total of 446 alleles were detected using 40 SSR markers while 11.2 alleles was detected on average per locus with the range of 3-27, expected heterozygosity was 0.742 on average with the range of 0.493-0.901 and polymorphism information content (PIC) was 0.707 with the range of 0.493-0.901. These parameters indicated high genetic information in Chinese and American alfalfa cultivar resources. For the high genetic diversity displayed by mtic238, mtic188, bf111, afctt1, bf641851, maa660456, aw361, etc., they were suitable to disclose the genetic diversity in Chinese and American alfalfa. In terms of different chromosomes, the second and eighth chromosomes possessed relatively high genetic diversity while the first relatively low. In the analysis of population structure based on the admixture model, two different methods determined the optimal K was 2, which meant 100 genotypes from 16 cultivars could be grouped into two sub-populations even with certain discrepancy. The principle component analysis and clustering analysis were consistent with the analysis of population structure. The genetic diversity of Chinese cultivars was slightly higher than American’s, but not significant. 【Conclusion】 The germplasms from China and the United States were relatively highly diversified and variable in genetic analyses, revealing high genetic diversity. Difference exists in the diversity of genotypes of both countries: Chinese germplasm is slightly higher than those from America. The fact that the population structure of all genotypes was not strictly consistent with the original countries is credited to the allogamy and wide genetic exchange in alfalfa.

Key words: alfalfa , tetraploid , SSR marker , genetic diversity , population structure