Abstract: We aimed to establish the high throughput sequencing method for soil microbial 16S rDNA using Ion Torrent PGM. In this study， the V4\|V5 variable region of 16s rDNA was amplified from the marsh soil and meadow soil by PCR and sequenced respectively. The Shannon curves of the two soil samples tended to be stabilized when the sequencing depth reached 15 000 sequences. The diversity index showed that the bacterial communities in the marsh soil were more abundant than that in the meadow soil. A total of 7 615 and 8 072 OTUs were identified in the marsh soil and meadow soil， respectively， which corresponded to 30 phyla，53 classes，93 orders，130 families and 182 genera in the marsh soil and 29 phyla，52 classes，88 orders，125 families and 183 genera in the meadow soil. The variance analysis showed that there were significant differences in the abundance of microbial compositions and OTUs between the marsh soil and meadow soil. According to the similarity analysis， all the tested samples could be divided into two groups， and the three repeats were classified into the same group. Therefore，it was suggested that the soil microbial 16S rDNA detection method based on Ion Torrent PGM developed in our study could comprehensively and accurately reflect the soil microbial community structure， and provide a powerful tool for the detection of soil microbial community structure.

Key words: Ion Torrent PGM, soil microbial community, 16S rDNA