Fig. S1 Number of raw reads obtained per individual for each of the 29 species in this analysis.

Fig. S2 Coverage obtained per individual. Due to the low number of reads achieved for X. evelynae, individual samples for this species were combined for further analyses.

Fig. S3 SNP matrix size found when different numbers of maximum mismatches (5 or 8) and minimum number of species (15, 20 or 25) are implemented.

Fig. S4 Phylogenetic estimations using ML with different maximum mismatches and minimum number of species (the following order is shown at each node when estimations differed from bootstrap values of 100—m5, sp15; m5, sp20; m5, sp25; m8, sp15; m8, sp20; m8, sp25). *indicates a different topology at that node; in the clemenciae clade, X. monticolus and X. clemenciae were found to be the most derived species using one of six matrices. In the northern platyfish clade, X. couchianus and X. gordoni were found to be the most derived using two of six matrices. These estimations were less well supported than the topology found using the majority of the SNP matrices.

Table S1 Overview of RAD raw data and experimental setup. Individual raw read data, read data retained after quality control, library number (multiplexing design) and sequencing platform.

Table S2 Proportion of missing data at all parameters analysed in this study where maximum and minimum numbers are calculated for the ingroup taxa (see also Fig. S4).

Table S3 Estimated time of origin of the major Xiphophorus clades including mean and 95% confidence interval values (see also Methods).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.