Sapindus saponaria L.is an economic tree species for biodiesel,medical,ecological and many other uses.In this study,MISA software was used to mine SSRs in transcriptome of S.saponaria.The SSRs characteristics were analyzed.SSR-PCR reaction system were constructed and optimized.Results showed 52460 unigenes in S.saponaria transcriptome.Among which 7014 unigenes contained SSRs,and a total of 8654 SSRs were discovered.SSR richness was decreased from dinucleotide to hexanucleotide.SSR length ranged from 12 to 250 bp.3619 SSRs had the length less than or equal to 15 bp,all of which were dinucleotide and trinucleotide.The length of the SSRs was greater than 15 bp,mainly dinucleotides and trinucleotides.Totally 435 different repeat units were in these SSRs,including 139,115,109,60,12 and 0 types respectively,from hexanucleotide to mononucleotide.The SSR repeat elements frequencies between 4 to 30 times,89.65%of the total SSRs accounted to 7090 locis were distribut-ed from 4 to 10 times.GO functional annotation of target genes at SSR locis showed that the Unigenes could be di-vided into 3 categories and 51 groups.KEGG function annotation showed that the metabolism pathway accounted for 65.18%in 5 categories.Through the construction and optimization of the reaction system of S.saponaria SSR-PCR,the optimal combination was as follows:37 cycles,annealing temperature 60℃,primer amount 2.0 µL and DNA template concentration 40 ng/µL.