中文摘要：

中国的天山及其附近山脉是世界拟南芥及其近缘种的主要分布区之一。在对天山—阿尔泰山的浅山地带拟南芥生存分布多年调查的基础上,选取13个代表性样地及18个相关环境因子,研究了拟南芥种群生存分布与环境因子的关系。结果表明：双向指示种分析（TWINSPAN）将13个拟南芥样地分为4个群落类型：新疆绢蒿、新疆绢蒿—猪毛菜、新疆绢蒿—刺叶锦鸡儿—草原苔草、勿忘草—密穗雀麦—草原苔草4类,各类型对应的环境因子异质性明显。不同群落类型在去势对应分析（DCA）、主成分分析（PCA）和典范对应分析（CCA）各排序图中区分明显,与物种对应较好,与TWINSPAN分类结果基本一致。环境因子的PCA分析发现,坡向、有机质、电导率、pH、土壤含水量和有效钾（均为2个土层）是导致样地间环境异质的主要因素,且这些因子间多呈显著关联。CCA分析发现坡向、有机质、pH、电导率和土壤含水量（均为2个土层）与排序轴相关性最大,分析认为上述环境因子是决定物种分布及多样性格局的主导因子,与环境因子的PCA分析结果基本一致。研究还发现,坡向、土壤含水量（第2层）、有机质（第1层）和pH（两层）是影响各样地十字花科种数变化的主要因素,电导率（第1层）是影响拟南芥数量分布的主要因素。

英文摘要：

Tianshan Mountains and the nearby mountains are one of the main distribution regions of Arabidopsis thaliana and its closely related species in the world. Based on comprehensive field surveys on the distribution of A. thaliana in the Tianshan-Altay mountain region, 13 plots in low mountain zone Shihezi, the 143rd Corps, Shawan, Dushanzi, Yili Guozigou, Emin and Ahay were seleeted and 18 environmental factors were measured, the relationship between A. thaliana populations distribution and environmental faetors were studied. Classifieation of the vegetation was analyzed using two-way indicator species analysis （TWINSPAN） technique. Ordination techniques as detrended correspondence analysis （DCA）, principal component analysis （PCA） and canonical correspondence analysis （CCA） were used to examine the relationships between vegetation and environmental parameters. Lastly, the correlation coefficients between cruciferous species, A. thaliana and environmental variables were also examined. The results showed that 13 plots were classified into 4 community types by TWINSPAN using important values （IV） data of 53 main species： Seriphidium kaschgaricum, Seriphidium kaschgaricum-Salsola collina, S.kaschgaricum- Cara acanthophylla-Carex liparocarpos, and Myosotis sylvatica-Bromus sewerzowii-C, liparocarpos. And environmental heterogeneity obviously appeared between each community type. Distribupos. And environmental heterogeneity obviously appeared between each community type. Distributions of community types in DCA, PCA and CCA ordination figures showed obvious distinction, and they corresponded to species distribution well. The result was basically the same as TWINSPAN classification. PCA analysis on environmental factors showed that slope, organic matter, electric conductivity, pH, soil moisture and available potassium （both 2 soil layers） were the main factors affecting environmental heterogeneity in different plots, and these factors were most significantly associated with each other. CCA analysi