1 N, number of individuals analyzed; n, number of haplotypes; h, haplotype diversity; and π, nuclear diversity 2.2 DNA extraction, polymerase chain reaction (PCR) amplification, and sequencing Total genomic DNA was isolated from frozen samples. The COI was amplified with the primers COII8068F(5’-CCATAACAGGACTAGCAGCATC-3’) and COIOCTR (5’-ATCATAGCATAGACCATACC-3’)(McFadden et al., 2006). All the reactions were performed in 50 μL volume total volume, with 1 μL DNA extract, 5 μL 10× buffer (2.5 mM Mg2+), 1 μL Taq DNA polymerase (5 U/μL), 1 μL dNTP (10mM), 1.5 μL F-primer (10 μM), 1.5 μL R-primer (10uM), and 39 μL ddH2O. The amplification protocol for COI and ITS consisted of 5 min of initial denaturation at 95 °C followed by 35 cycles of 30 s at 95 °C, annealing at 58 °C for 30 s, extension at 72 °C for 90 s and a final extension at 72 °C for 7 min. The resulting PCR products were purified and sequenced using an ABI 3730XL Genetic Analyzer (Applied Biosystems) with the same forward and reverse PCR primers. The initial sequences were assembled using SeqMan and aligned and trimmed using BioEdit. 2.3 Phylogenetic analyses The number of haplotypes for each locus was determined using DnaSP software. The number of polymorphic sites, haplotype diversity, and nucleotide diversity were estimated using the Arlequin(Excoffier et al., 2005). The distribution of characteristic nucleotide mutations in each haplotype was calculated using MEGA 6.06(Kumar et al., 2004). Bayesian inference (BI) and maximum likelihood (ML) trees were constructed using BEAST v.2.6.7 and MEGA 6.06, respectively. 2.4 Population analyses Analysis of molecular variance (AMOVA) was performed for the COI data using Arlequin with 1000 permutations to examine the amount of genetic diversity partitioned among populations(Excoffier et al., 1992). In addition, pairwise genetic differentiation (FST) among populations was calculated using the Arlequin software. Mantel’s test with 10000 randomizations for isolation-by-distance (IBD) was performed between the linearized FST(FST/(1-FST)) and geographical distances(Slatkin, 1995) using IBD v 1.52 (Bohonak, 2002) to determine whether this pattern meets the expectation of genetic similarity decaying with geographic distance(Novembre et al., 2008). 2.5 Historical demography analyses Extended Bayesian Skyline Plots (EBSPs) were used to examine population size changes during BEAST. Tajima’s D(Tajima, 1989) and Fu’s Fs(Fu, 1997) neutrality tests were performed using Arlequin, and a mismatch distribution was constructed for each geographic population to test the exponential population growth model. 3. Results 3.1. Seqence variation An alignment of the 892 bp COI gene fragment was analyzed for 71 individuals, with 494 variable sites yielding a total of six unique haplotypes named H1-H6 (Figure 1). The central haplotype, H1, was the most common, with 64 copies, and was dominant at all five sites. The genetic variation of COI mtDNA in the five populations of C. obesa from the ECS was low, with haplotype diversity ranging from 0.000 in SH to 0.600 in NT, and nucleotide diversity ranging from 0.0000 in SH to 0.1266 in NJ (Table 1). 3.2. Population sctructure The AMOVA results for COI showed that the differentiation among populations was not significant (P > 0.05) (Table 2), with FST values ranging -0.0774–0.1532. IBD analysis also revealed a non-significant (P > 0.05) correlation between genetic and geographic distances for all populations in the ECS. The ML tree showed that most of the C. obesa clustered into one clade (Figure 2), with the exception of a few individuals.Table 2. AMOVA comparing genetic variation based on COI sequences of C. obesa