Abstract: The transcriptome sequences of Diodon hystrix were obtained by RNA-seq technology, and a total of 221 762 Unigenes were generated by de novo assembly with N50 of 2 240 nt and GC content of 46.20%. By using MISA software, 106 221 SSR loci, which distributed in 62 451 Unigenes with a frequency of 28.16%, were detected from the RNA-seq data of D. hystrix. The dominant repeat units were mononucleotide, dinucleotide and trinucleotide, which accounted for 48.99%, 32.57% and 14.72% of the total SSR loci, respectively. The A/T was the main repeat unit in mononucleotide, accounting for 46.21% of the total SSR loci, while the AC/GT and AGG/CCT were the dominant repeat units in di- and thinucleotides, accounting for 21.90% and 2.70% of the total SSR loci, respectively. Altogether 17 563 pairs of primers were designed by selecting SSR loci with length greater than 20 bp. Then 160 pairs of primers were randomly selected for amplification and identification, and 95 pairs of effective amplification primers were screened, accounting for 59.38%. Thirty pairs of stable and repeatable polymorphic primers were obtained from the effective amplification primers by polymorphism verification (31.58% of the effective amplification primers). Among them, 15 pairs of primers showed high polymorphism (PIC>0.5), which were benefit for assessing the diversity of D. hystrix population. These results indicate that the transcriptome data of D. hystrix can be used as an effective source for the development of stable SSR markers, and the obtained polymorphic SSR loci can provide foundation for the further study of genetic map and genetic diversity of D. hystrix.