Abstract: Association mapping is a procedure for detecting QTLs as well as their alleles based on linkage disequilibrium (LD). The genotyping data of 60 simple-sequence repeat (SSR) markers on representative samples of 393 landraces of Glycine max (L.) Merr. and 196 wild accessions of Glycine soja Sieb. et Zucc. were used in the present study. Linkage disequilibrium of pairwise loci and population structure were firstly analyzed for the two populations, then the association analysis between SSR loci and 16 agronomic and quality traits was performed by using TASSEL GLM (general linear model) program. The different degrees of LD were detected not only among syntenic markers but also among nonsyntenic ones, suggesting that there had been historical re-combination among linkage groups. The cultivated soybean population had more LD loci pairs than wild soybean population, while the later had higher degree and slower attenuation of LD than the former. Genetic structure analysis showed that both of the cultivated and wild populations were composed of nine and four subpopulations, respectively, which associated with their geo-graphic eco-types, indicating the classification of geographic eco-types was of sound genetic bases. Twenty seven and thirty four SSR loci associated with the traits were screened out from cultivated and wild populations, respectively. Some loci were found to associate with a same trait in both populations, and there existed both consistent and inconsistent association between the culti-vated and wild populations. There were a few loci associated with two or more traits simultaneously, which might be the genetic reason of correlation among traits or pleiotropic phenomena. In addition, twenty-four associated markers were in agreement with mapped QTLs from family-based linkage mapping procedure.

Key words: Cultivated soybean [Glycine max (L.) Merr.], Wild soybean (Glycine soja Sieb. et Zucc.), Simple-sequence repeat (SSR), Population structure, Association mapping