Abstract: In order to explore the feasibility of using the feature elements between origins to identify origins,the contents of 10 elements in Ziziphus jujubas which collected from different producing areas were measured,and the data were analyzed by difference analysis,cluster analysis,Fisher discriminant analysis and partial least squares discriminant analysis( PLS- DA).The results showed that there were significant difference in the contents of Mg,B,Mn,Fe,Zn and other elements from the two places of origin,and R- type hierarchical cluster analysis of 10 elements indicated that B,Mn,Fe and Zn element clustered together with similar behaviors.The correct classification rate of geographic origin made by Fisher discriminant analysis,Q- type hierarchical cluster analysis and PLS- DA based on feature elements were closed to the result of total elements.The highest classification rate was made by PLS- DA based on feature elements,which gave an overall correct classification rate of 94.0% and cross- validation rate of 94.0%.The correct rate of Q- type hierarchical cluster analysis was lowest,which was only84.06%.This study confirmed that the feature elements between origins was effective in predictions of geographic origin,and supervised modes Fisher discriminant analysis and PLS- DA were superior to unsupervised mode Q- type hierarchical cluster analysis at origin traceability.