Abstract: 【Objective】In order to rapidly and precisely identify the excellent gene for the breeding of Eucalyptus urophylla, the construction of core collection of E.urophylla was necessary. Establishing a core collection of E.urophylla could simplify the management and enhance the utilization of genetic resources in E.urophylla research programs. 【Method】Based on 17 quantitative traits, height, diameter and volume of 194 families from 27 provenances of E.urophylla, core collections were constructed using different sampling proportions (5%, 10%, 20%, 30%, and 40%), 3 distances (Euclidean distance, Standardized Euclidean distance, and Mahalanobis distance), 4 hierarchical clustering methods (Single distance method, Complete distance method, Unweighted pair-group average method and Ward’s method), and 2 sampling strategies (random sampling and least distance stepwise sampling) were employed. The genetic variation of the quantitative traits among these core collections was compared by evaluating the mean difference percentage (MD), variance difference percentage (VD), changeable rate of coefficient of variation (VR) and coincidence rate of range (CR) of the traits.【Result and conclusion】The results showed that the primary core collection was constructed with 10% sampling proportion, standardized Euclidean distance, Single distance method,and least distance stepwise sampling, which was composed of 100 families from 27 provenances of E.urophylla and the best representative of total germplasm.