中文摘要：

长序榆是我国二级重点保护的濒危植物，对研究榆属、榆科植物的系统发生有重要意义，并且具有潜在的经济价值。研究长序榆的生境特点，对该种的引种、科学保护有重大意义，通过对分布在我国的长序榆近两年的踏查发现，长序榆主要集中在我国的安徽、浙江、江西和福建等省，长序榆生境已经严重片段化，自然干扰和人为干扰是造成其濒临灭绝的主要原因。长序榆多分布在海拔600-900m的阳坡或半阳坡上，其中最大的种群分布在浙江开化和遂昌，分布区的特殊地势使其成为长序榆的冰期避难所。对18项生物学、生态学指标的主成分分析结果表明，影响长序榆生存的主要环境因子是：光照、土壤养分、坡向和海拔。对浙江松阳、江西武宁和福建南平数量极少的种群需优先进行迁地保护，积极引种栽培；而对于安徽歙县、浙江临安、开化和遂昌等大种群分布区，应该扩大核心保护区面积，避免人为破坏。

英文摘要：

The Grade Ⅱ protected endangered plant Ulmus elongata （Ulmaceae） is a significant phylogenetic member of the genus and the family; it may also have economic value. Research on the habitats occupied by U. elongata is important in support of its potential introduction into horticulture and for the conservation of this species. Ulmus elongata occurs mainly in Anhui, Fujian, Jiangxi and Zhejiang provinces based on our two years of field surveys in China. We concluded habitats of U. elongata have become severely fragmented. The main causes of local extirpations of this species were both natural events and human activities. The main forms of human impact were habitat destruction caused by mining and tourism, the demand for firewood from mature trees by local farmers and the urbanization and development of forest plantations. Natural disturbances were primarily caused by extreme weather, such as the blizzards of 2008, which caused extensive damage to U. elongata populations in Suichang County, Zhejiang Province, but natural disturbances cause less damage than human activities, Ulmus elongata primarily occurs at altitudes of 600--900m on sunny or partly sunny slopes. Suichang and Kaihua County in Zhejiang Province were glacial refugia and had the Spearman correlation test shows the correlations of 18 biological and largest populations because of their special terrain. A environmental factors to each other were significant （P〉0.05）, but only slightly or barely significant because of serious habitat fragmentation, indicating the communities of U. elongata were at different stages of succession. Principal component analysis shows the most important biological and environmental factors of habitats were light, soil nutrients, aspect and elevation. We extracted 4 principal components. Principal component 1, representing a light and energy factor, explained 28. 408% of the total variance; principal component 2, driven by a soil factor, explained 27. 031% of the variance; principal component 3, slope, accounted for