Nucleotide sequences were edited in MEGA5 (Tamura et al., 2011) or MacClade 4.08a (Maddison and Maddison, 2000). The DNA sequences were aligned using Clustal W (Larkin et al., 2007) or manually. Heterozygous sites within individual specimens were rare (0–1.1% for genes); these were coded as IUPAC degenerate nucleotide symbols and directly used in the analyses.The obtained sequences were deposited in DDBJ/GenBank/EMBL (accession nos.AB988263–AB988988).