3.4. Sequence analysis

The sequences were aligned using programs PILEUP and LINEUP in Genetic Computer Group Program Package (available in CSC) and ClustalW (Thompson et al. 1994), when needed. All alignments were checked and edited manually. The pairwise genetic distances were estimated using Kimura’s two-parameter method (Kimura 1980), HKY85 (Hasekawa et al. 1985), F81 (Felsenstein 1981), TAMNEI (Tamura & Nei 1993), TVM (Rodriquez et al. 1990) with or without gamma distribution parameter. In I, Modeltest program (Posada & Crandall 1998) was used to choose the model of sequence evolution that best fits the data. Neighbor-joining method in MEGA (Saitou & Nei 1987, Kumar et al. 1993), maximum parsimony method in PAUP3.1 (Swofford 1993), maximum likelihood method in PAUP*4.0b6 (Swofford 1998) and maximum likelihood method in fastDNAml (Olsen et al. 1994) were used for inferring the phylogenetic relationships of the taxa or mtDNA haplotypes.

At the intraspecific level, haplotype (^h) and nucleotide (π ) diversity were calculated as in Nei (1987, eqs. 8.5 and 10.5, respectively). The amount of genetic differentiation among the localities was estimated by F_ST (based on haplotype frequencies) and &phis;_ST (incorporating sequence variation) using AMOVA and their significance was tested with a randomisation procedure in AMOVA (Excoffier et al. 1992). The amount of female gene flow was estimated based on population pairwise F_ST values using the equation N_fm=1/2(1/F_ST-1) (Nei 1987, eq. 13.25).