COI gene analyses
To detect the COI haplotypic structure of A. alalia and A. mantiqueira , a haplotype network was designed using PopART with the TCS network option (Leigh & Bryant, 2015). To propose a phylogenetic hypothesis, a Bayesian inference (BI) was carried out using MrBayes v. 3.2 (Ronquist et al., 2012) on the CIPRES portal (Miller et al., 2010), including one individual of A. catarina as an outgroup (Table 1). The model-jumping feature was employed to select the best nucleotide substitution model (Ronquist & Huelsenbeck, 2003). The model with the highest posterior probability was the GTR submodel [121131]. The gamma parameter was also included to allow site rate variation. Four simultaneous chains were run for 107 generations in two runs, sampling trees every 1,000 cycles. The first 25% of trees were discarded as “burn-in”. The convergence of the likelihood traces of the independent runs was assessed with TRACER v1.5 (Rambaut et al., 2018), and the ESS (effective sample size) values were verified to be above 3,700 for all parameters, which indicates that they were sufficiently sampled to estimate their posterior distributions (Drummond et al., 2006).