Figure S1. NeighborNet network from SpliTstree4 showing reticulate relationships among lineages.

Figure S2. Marginal posterior probability for (A) current Ne and ancestral Ne, and (B) asymmetric population migration rates (2Nem) estimated with IMa2.

Figure S3. Boxplots illustrating the observed differences in (A) herkogamy, (B) protandry, (C) flower size, and (D) autofertility xantiana and parviflora.

Figure S4. Ancestral character reconstructions based on the best topology from GARLI (Fig. 2) for (A) protandry, (B) flower size, (C) autofertility, and (D) herkogamy.

Table S1. Population codes (Site #), location, number of samples (N), elevation, and geographic coordinates of individuals included in this study.

Table S2. Population-level estimates of θw and Tajima's D.

Table S3. Taxon-level estimates of θw and Tajima's D.

EVO_1521_sm_SuppMatf1.pdf384KSupporting info item
EVO_1521_sm_SuppMatf2.pdf810KSupporting info item
EVO_1521_sm_SuppMatf3.pdf220KSupporting info item
EVO_1521_sm_SuppMatf4.pdf349KSupporting info item
EVO_1521_sm_SuppMatt1.pdf48KSupporting info item
EVO_1521_sm_SuppMatt2.pdf81KSupporting info item
EVO_1521_sm_SuppMatt3.pdf47KSupporting info item

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.