Soil microbial communities are a key driver in biogeochemical cycling and play an important role in ecosystem carbon and nitrogen cycling. However, microbial responses to climate change remain poorly understood. Mountain slopes provide a natural laboratory for studies of soil microbial diversity and biogeography. Understanding patterns of soil microbial composition, the distribution along elevational gradients and the factors driving such patterns is indispensable to achieve a comprehensive understanding of the response of ecosystems to global climate change. The Qilian Mountains on the northeastern Qinghai-Tibetan Plateau (QTP) represent one of the main areas of alpine mountains in China. In addition,the mountains are located at the juncture of three climate regions of western China (monsoon, arid and QTP climates). The study area, located in the mid part of the Qilian Mountains, is one of the regions with continuous alpine meadow soil distribution in the world. Soil samples were collected from the shady slope of Qilian Mountains. Topographic and climatic variations result in the vertical zonation of major vegetation types. In this study, we selected four elevation gradients, representing four typical vegetation types based on their elevation. The elevations 3200-3500 m are alpine shrub, dominated by Potentilla fruticosa and Salix cupularis. The elevations 3500-3700 m are swamp meadow, dominated by Kobresia humilis and K. capillifolia. The elevations 3700-3900 m are alpine meadow, dominated by K. tibetica,K. humilis and K. pygmaea. The elevations 3900-4200 m are cushion plants, dominated by Thylacospermum caespitosum and Androsace tapete and Saussurea medusa. Between 3300 and 4200 m, we sampled across the four vegetation types described above at elevation intervals of 300 m. At each elevation,soil samples were collected from three plots (20 m x 20 m) as three independent replicates. In each plot, soil samples of the surface layer (0-20 cm) were collected at five random points and bulked together as a single sample. High-throughput sequencing is revolutionizing microbial ecology studies, because it can obtain the classification information for soil bacteria more accurately than previous methods. In July 2013, we investigated soil bacterial diversity under the four typical vegetation types along a 3300-4200 m elevation gradient using Illumina high-throughput sequencing. We found that the soil microbial biomass carbon and nitrogen varied in the order: swamp meadow > alpine meadow > alpine shrub > cushion plants. In addition, across all soil samples, we obtained 713, 386 quality sequences with 54161-67678 sequences per sample (mean 59449). The dominant phyla across all the soils were Actinobacteria (23.10%), Acidobacteria (18.90%), Alphaproteobacteria (10.59%), Firmicutes (8.13%), and Gemmatimonadetes (6.12%),accounting for more than 63% of the bacterial sequences from each of the soils. Finally, soil pH, moisture,soil organic carbon and total nitrogen were correlated with the bacterial community composition (Mantel tests) and bacterial alpha diversity (operational taxonomic units and phylogenetic diversity). The results of the canonical correspondence and Pearson correlation analysis showed that soil pH,moisture, soil organic carbon and total nitrogen were the main factors that significantly affected the soil bacterial community composition and alpha diversity in these four vegetation types. This research provides a scientific basis for the ecosystem stability and protection of the Qilian Mountains. Further work is needed to understand the underlying mechanisms and ecological processes behind soil microbial composition and diversity.