Appendix S1 Locality and sampling information for molecular and morphological data in the C.  striata complex.

Appendix S2 Maximum Likelihood gene tree for combined rbcL and rpl32-trnL spacer from RAxML under the GTRGAMMA model.

Appendix S3 A. Tests of recombination for nuclear intron loci. RM = minimum number of recombination events based on Hudson and Kaplan (1985); Φ(P) = significance of the Φ-test for recombination (Bruen, 2006) implemented in SplitsTree v.4 (Huson and Bryant, 2006). B. Pairwise ФCT estimates of differentiation based on nuclear intron loci, partitioned by plastid DNA clade. ***P < 0.0001. C. Heterozygosity values and Hardy-Weinberg equilibrium tests (10,000 permutations) for nuclear intron loci in C.  striata s.s., partitioned by plastid clade. Symbols HO and HE signify observed and expected heterozygosities, respectively. D. Molecular diversity parameters for C.  striata s.s. nuclear intron and concatenated plastid loci, partitioned by plastid grouping. S = number of segregating sites, h = number of alleles, Hd = haplotype diversity, π = nucleotide diversity, θS and θπ = scaled mutation rate estimates based on # segregating sites and pairwise differences per site, respectively. ‘Total/mean’ represents nuclear DNA totals for S and h, and per-site averages across loci for the remaining parameters. Tajima’s D and Fu’s FS were non-significant in all cases. Sample sizes are in parentheses next to plastid clade grouping names (nuclear, plastid). Nuclear was ITS not included due to limited within-group variation. E. Nuclear gene flow and effective population size estimates from Bayesian coalescent analyses in MIGRATE, with 95% credibility interval (2.50–97.50% CI). Parameter conversions for effective population sizes and numbers of effective migrants per generation were Ne = Θ/4 μ and Nemi[RIGHTWARDS ARROW]j = ΘiM i[RIGHTWARDS ARROW]j/4, respectively. F. Character coefficients, eigenvalues, and percent variation explained for the first and second Canonical Variates (CV1, CV2) and Principal Components (PC1, PC2) for combined analysis of DNA (nominal-categorical scale) and floral measurement data (continuous scale). See Fig. 6 for CVA and PCA plots. CV coefficients are non-standardized, and PC coefficients are based on the correlation matrix. G. Landmark characters scored for the C.  striata complex: basal labellum ridge, labellum widest point, labellum apex, basal callus groove, basal callus ridge, callus widest point, apical callus ridge, and apical callus groove. H. Bayesian species delimitation using BPP 2.0 for the C.  striata complex under various gamma priors for Θ and τ0. The gamma prior G(α,β) has mean α/β and variance α/β2 (Yang & Rannala 2010). Primary guide tree topology is based on the plastid DNA analysis of Barrett & Freudenstein (2009): (((vreelandii, striata), California), (bentleyi, involuta)). Posterior probability values below are averages between two independently seeded runs. Model 1111 represents the full five species model illustrated by the guide tree, while model 1110 represents the four species model in which C.  bentleyi and C.  involuta are collapsed into a single species.

Appendix S4 MP phylograms of the C.  striata complex based on nuclear data.

Appendix S5 Population structure of Corallorhiza striata s.s.

Appendix S6 Posterior densities of migration estimates from Bayesian analysis in MIGRATE for C.  striata s.s.

MEC_5124_sm_OnlineAppendices.pdf293KSupporting 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.