应用幼苗萌发法对比研究了我国北方农牧交错区克氏针茅（Stipa krylovii）草原在放牧和围封条件下土壤种子库的差异。围封样地的土壤种子库中记录有16个物种，其中禾草类6种，非禾草类10种；若以生活型划分，则全部为多年生植物；若以单子叶或双子叶植物 划分，则单子叶植物9种，双子叶植物7种。放牧样地的土壤种子库中记录有13个物种，其中禾草类4种，非禾草类9种；多年生植物12种，一二年生植物1种；单子叶植物5种，双子叶植物8种。放牧样地土壤种子库的密度为(3 664±1 087)粒•m－2，围封样地土壤种子 库的密度是(5 139±1 848)粒•m-2，差异显著。采用Sorensen指数计算土壤种子库与地上植被的物种相似性，围封样地和放牧样地分别是0.711 1和0.611 1。两个地点的全部种子中，81.0%属于冷蒿（Artemisia frigida）、克氏针茅、星毛委陵菜（Potentilla acaulis）、糙隐子草（Cleistogenes squarrosa）和砂韭（Allium bidentatum）5种优势植物，16.2% 属于二裂委陵菜（P. bifurca）、菊叶委陵菜（P. tanacetifolia）、猪毛蒿（A. scoparia） 、冰草（Agropyron cristatum）和羊草（Leymus chinensis）5种常见植物，而其余9种植物的种子数只占2.8%。此外，通常作为优良牧草的豆科植物的种子数仅占0.6%。上述结果表明 ，在放牧条件下由于土壤种子库中一些重要物种的缺失或数量很少以及极不均匀的分布，可能会降低退化克氏针茅草原的自然恢复速度。我们建议采取补播这些重要物种并结合其它恢复措施来加快退化克氏针茅草原的恢复进程。
We studied the soil seed bank of a Stipa krylovii steppe community at two sites with different land_use histories: an enclosed pasture, which has been fenced since 2001, and a grazed pasture, which has been continuously grazed for more than 20 years. The experiment was carried out in Duolun County of Xilingol League, Inner Mongolia. Sixteen 12 cm diameter soil cores were collected to a depth of 5 cm at each site on March 30 and 31, 2004. The soil seed bank was estimated by monitoring seedling emergence, a convenient way to compare soils under similar conditions. Vegetation composition was assessed at each site on May and August 2004 by recording the species presented in sixteen 0.5 m×0.5 m quadrats that were distributed randomly alongside the seed bank sampling points. A total of 1 593 seedlings belonging to 19 species germinated from 32 soil samples. The total seed bank flora at the two sites included 6 grass species and 13 forbs, 18 of which were perennials and 1 annual, 10 monocotyledons and 9 dicotyledons. In the fenced pasture, there were 16 perennials, 6 grasses and 9 monocotyledons from a total of 16 species in the seed bank, whereas 12 perennials, 4 grasses and 5 monocotyledons from a total of 13 species were found in the seed bank of the grazed pasture. The average number of viable seeds per square meter was s ignificantly higher at the fenced site A ((5 139±1 848) seeds•m－2) than that at the grazed site B ((3 664±1 087) seeds•m－2). Sorensen‘s similarity index was 0.711 1 for the fenced site and 0.611 1 for the grazed site. In total, only fifteen species were found in the established vegetation at both sites. As many as 81.0 % of the seeds recorded belonged to only five species (Artemisia frigida, S. krylovii, Potentilla acaulis, Cleistogenes squarrosa, and Allium bidentatum). Another 5 species accounted for 16.2%of the seeds recorded and the other 9 species accounted for only 2.8% of the total. Our results showed that the scarcity of seeds of some important steppe species combined with the unbalanced distribution of seeds among species may inhibit the restoration process of the S. krylovii steppe. Therefore, reseeding and other management steps should be used to speed up the restoration process of the degraded S. krylovii steppe.