Abstract: Mungbean (Vigna radiate) is planted extensively in China, and there are abundant Chinese mungbean germplasm resources. In order to strengthen the preservation, research and utilization of Chinese mungbean germplasm resources, we established the pre-core collections of mungbean germplasm. In this study, 5 072 mungbean germplasm resources were selected as material. Based on 14 agronomic traits, 13 sampling strategies were used, which consisted of 2 stratification methods, 3 sampling number determination methods and 2 individual selection methods. The criteria against whole collection included mean similarity coefficient, coincidence of traits, mean CV of quantitive traits, mean Shannon index. The results indicated that selection based on clustering was better than random selection, stratification according to traits was better than that according to geographical distribution, genetic diversity-dependent strategy was better than square root and proportion strategies in determination of sampling number. Finally, an optimal strategy was set by stratification according to traits, determination of the sampling number with genetic diversity-dependent strategy, and then selecting sample by clusters. By using this strategy, a pre-core collection of Chinese mungbean germplam with a total of 719 accessions was established, carrying 100% phenotypic genetic diversity of the whole collection.

Key words: Mungbean, Germplasm, Core collection