Abstract: Cunninghamia lanceolata is an important commercial timber and carbon sequestration tree species in south China. Based on the diameter at breast height (DBH) of 33 third-generation full-sib families of C. lanceolata from Guangdong at multiple tested sites and the multi-site ANOVAs approach, we identified a significant difference (P<0.01) of the genetic adaption in families. While, the site and block effects and the site×family interaction effect seemed to be profound. Regarding this, best linear unbiased prediction (BLUP) based on individual model was carried out to calculate the individual breeding value of the families in each experimental site. The obtained result highlighted the preliminarily separation of the families according to their adaptability and the superior families with higher individual breeding value and adaptability (e.g. families 1, 4, 13 and 31). Further genotype main effect plus genotype-by-environment interaction (GGE) biplot analysis directly gave ‘who won where’ functional diagram, high yield and stable families graph and ideal families selection result that highly agreed with that of BLUP’s, indicating a consistency of these two methods and their forecasting power.

Key words: Cunninghamia lanceolata, breeding values, BLUP, GGE, adaptation

摘要： 杉木Cunninghamia lanceolata是中国南方重要的商品用材和碳汇造林树种。本研究以广东杉木33个第三代全同胞家系在不同生长地点的8年生径向生长表现(胸径)为基础开展多地点方差分析。结果表明，家系适生性遗传差异大(P<0.01)，且地点、区组及地点与家系间的互作效应显著。采用基于个体模型的BLUP分析法计算每个试验地点参试家系的个体育种值，可将参试家系适生性进行初步划分，并筛选出个体育种值水平较高、适生性好的家系(家系1、4、13和31)。在此基础上，运用GGE双标图分析法进一步获得参试家系“哪个赢在哪里”功能图、不同家系高产性与稳产性结论、理想家系选择列表，发现家系1、4、13、31和其他家系适生性评价结果与BLUP分析结果基本一致，具有较好的分析效果。

关键词: 杉木, 育种值, BLUP, GGE, 适生性