中文摘要：

查阅了维普《中文期刊数据库》（1989--2006年）和Web of Science（1985--2006年）中发表的土壤种子库研究文献，按《中国植被》划分的29个植被类型对土壤种子库密度、丰富度和研究方法等数据进行归类总结，共采集到14个植被类型的238个样地信息．结果显示：研究者使用的研究方法及获得的种子库密度和丰富度数据差异巨大．所有研究中采样时间以4和10月最多；样方面积介于78～10000cm^2之间，样方数量介于2～480之间；10cm×10cm和20cm×20cm是最常用的采样方式；总采样面积介于600～500000cm^2之间，并以1000～10000cm^2为最多．土壤种子库密度值变化范围在8粒·m^2（沙漠）～65355粒·m^2（热带雨林的次生林）之间，物种数变化范围在1（温带草原次生光碱斑）～74（热带季雨林）之间，同一植被类型下的变异也相当大．热带的季雨林和雨林的密度值和物种数显著高于温带的针叶林；而人工林的密度和物种数量大于农地，农地又大于裸地．草原、荒漠和草甸的物种数量相对较少．未来的土壤种子库研究需要从广度和深度上进行扩展，并重点加强重要生态系统的土壤种子库长期定位研究以及种子库对策研究，特别是应将这些研究与植被群落的演替、更新和恢复有机地联系起来．同时应加强对不同植被类型的种子库采样、检测方法的探索性研究．

英文摘要：

By searching soil seecl bank （SSB） papers from http：//www. cqvip. corn （1989-2006） and Web of Science （1985-2006） , the information on SSB density, species richness, and research methods were summarized according to the 29 classified vegetation types in Vegetation of China. In total, the data of 238 sites with 14 vegetation types were collected. The results showed that the re- search methods adopted by different researchers and the obtained data of SSB density and species richness varied greatly. In related researches, sampling work was mostly conducted in April and October, sampling plot number ranged from 2 to 480, plot area ranged from 78 cm^2 to 10000 cm^2, with 10 cm × 10 cm and 20 cm × 20 cm as most common, and total sampling area ranged from 600 cm^2 to 500000 cm^2, with the most being 1000-10000 cm^2. SSB density varied from 8 ind · m^-2 （desert） to 65355 in d · m^-2（tropical rain forest）, and species richness varied from 1 （secondary bare alkali-saline patch in temperate） to 74 （tropical seasonal rain forest） per site. SSB density and species richness were higher in tropical rain forest and seasonal rain forest than in temperate coniferous forest, and in manmade forest than in agricultural land or barren land. Grassland, desert, and meadow had smaller species richness. In future, the SSB research should be extended both in scope and in deepness, with the focus on the long term research and strategy research of some important ecosystems, and the research should be incorporated into vegetation regeneration and restoration studies. The related methodological research should be also emphasized in the future.