为更好地保存、利用和评价落羽杉〔Taxodium distichum (Linn.) Rich.〕种质资源，以65份落羽杉为供试材料，对其11个定性性状和6个定量性状进行多样性、相关性和聚类分析，并筛选适宜的取样策略进行核心种质构建。结果表明：定性性状的Shannon-Wiener多样性指数为0.23~1.42，变异系数为20.48%~132.99%；定量性状的Shannon-Wiener多样性指数为1.94~2.06，变异系数为10.94%~20.00%。脱落性小枝长度与脱落性小枝宽度和针叶长度极显著正相关（P＜0.01），针叶密度与脱落性小枝长度、脱落性小枝宽度和针叶长度均极显著负相关，脱落性小枝宽度还与针叶长度和针叶夹角极显著正相关，针叶夹角还与枝密度显著负相关。聚类分析结果显示：65份落羽杉可分为4个类群，第Ⅰ类群的落羽杉多具有抗赤枯病、秋季叶色呈棕红色等性状；第Ⅱ类群表现为脱落性小枝下垂、针叶夹角较大、落叶期早等；第Ⅲ类群表现为脱落性小枝长度中等或较长、针叶不在同一平面、芽开放期晚等；第Ⅳ类群表现为枝密度和针叶密度较大等。筛选出“欧氏距离-优先取样法-可变类平均法-25%取样比例”为构建核心种质的最佳策略，采用此策略筛选出含16份落羽杉的核心种质，均值差异百分率、方差差异百分率、极差符合率和变异系数变化率分别为0.00%、17.65%、100.00%和119.88%，符合核心种质的构建要求；此外，核心种质获得了更大的变异，与原有种质有良好的异质性。综上所述，供试65份落羽杉材料的表型多样性较为丰富，筛选出的核心种质能够有效代表原有种质。

To better preserve, utilize, and evaluate germplasm resources of Taxodium distichum (Linn.) Rich., 65 T. distichum materials were taken as experimental materials, diversity, correlation, and clustering analyses were conducted for 11 qualitative traits and 6 quantitative traits, and appropriate sampling strategies were screened for core collection construction. The results show that the Shannon-Wiener diversity indexes of qualitative traits are 0.23-1.42, and the coefficients of variation are 20.48%-132.99%; the Shannon-Wiener diversity indexes of quantitative traits are 1.94-2.06, and the coefficients of variation are 10.94%-20.00%. The exfoliating branchlet length shows extremely significant positive correlations (P<0.01) with exfoliating branchlet width and needle length, while the needle density shows extremely significant negative correlations with exfoliating branchlet length, exfoliating branchlet width, and needle length, and the exfoliating branchlet width also shows extremely significant positive correlations with needle length and needle angle, whereas the needle angle has a significant negative correlation with branch density. The clustering analysis result shows that the 65 T. distichum materials can be divided into four groups, group Ⅰ of T. distichum mostly have traits including resistance to red blight disease, brownishred leaf color in autumn, etc.; group Ⅱ displays exfoliating branchlet droop, relatively large needle angles, early defoliation periods, etc.; group Ⅲ has medium or relatively long exfoliating branchlet length, needles not in the same plane, late bud opening period, etc.; group Ⅳ shows relatively large branch density and needle density, etc. The optimal strategy for core collection construction is identified as ‘Euclidean distance-priority sampling method-variable average method-25% sampling proportion’, a core collection of 16 T. distichum materials are screened out by using this strategy, and the mean difference percentage, variance difference percentage, range coincidence rate, and coefficient of variation change rate are 0.00%, 17.65%, 100.00%, and 119.88%, respectively, meeting core collection construction criteria; in addition, the core collection has obtained greater variability, and maintains good heterogeneity compared to the original collection. In conclusion, the phenotypic diversity of the 65 T. distichum test materials is relatively rich, and the core collection screened out can effectively represent the original collection.