Table S1.Information for the microsatellite loci used in this study, this includes: locus name, original reference for development of that locus, GenBank accession number (if available), primer specific annealing temperature (°C), concentration [μM] of forward and reverse primers, the multiplex PCR set for amplification, and multiplex genotyping set for each microsatellite locus.

Table S2. For each location, data is shown for the number of specimens analyzed with each microsatellite locus (n), FIS values used to detect deviation from the Hardy–Weinberg equilibrium, observed (HO) and expected (HE) levels of heterozygosity, and allelic richness (AR). Significant p values are denoted as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Table S3. Null allele frequencies by location and microsatellite locus.Location abbreviations: JP, Japan; CA, California; MX-P, Mexico-Pacific; MX-G, Mexico-Gulf; CH, Chile; NZ, New Zealand; SA, South Africa.

Table S4. Parameters for BAYESASS replicates run to assess migration estimate convergence.

Data Set / Iterations / Burn-In / Seed #
Replicate 1 / 20000000 / 2000000 / 14000
Replicate 2 / 20000000 / 2000000 / 600000
Replicate 3 / 40000000 / 4000000 / 5
Replicate 4 / 40000000 / 4000000 / 40000
Replicate 5 / 80000000 / 8000000 / 637
Replicate 6 / 80000000 / 8000000 / 50000
Replicate 7 / 160000000 / 16000000 / 99999
Replicate 8 / 160000000 / 16000000 / 82
Replicate 9 / 100000000 / 10000000 / 7845
Replicate 10 / 100000000 / 10000000 / 589421

Table S5.Weir and Cockerham’s pairwise FST values (above diagonal) and Jost’spairwise DESTvalues (below diagonal) for null-correctedmicrosatellite data.Significant p values are denoted as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Location abbreviations: JP, Japan; CA, California; MX-P, Mexico-Pacific; MX-G, Mexico-Gulf; CH, Chile; NZ, New Zealand; SA, South Africa.

Table S6.Pairwise mtFST values (above the diagonal) based on mitochondrial sequence data, and DEST values based on microsatellite data (below the diagonal) for Seriolalalandi. Significant mtFST values (p ≤ 0.00238) are indicated by *; significant DESTpvalues are denoted as follows: *, p < 0.05; **, p < 0.01; ***, p < 0.001. Locations are abbreviated: JP, Japan; CA, California; MX-P, Mexico-Pacific; MX-G, Mexico-Gulf; CH, Chile; NZ, New Zealand; SA, South Africa. Sample sizes are included for each location for the microsatellite and mtDNA sequence analyses.

Table S7.Mitochondrial mismatch analyses.

a. b.

c. d.

Figure S1. Results from Bayesian clustering analyses using STRUCTURE. Plots (a) and (b) display mean log-likelihoodvalues (LnP(D)) for analyses using the admixture model without and with location prior, respectively, for 7 independent runs for each value of K for K =1–7. Plots (c) and (d) show the relationship of delta K to each value of K across the runs for analyses without and with location prior, respectively.

a. K = 2

b. K = 3

c. K = 4

Figure S2. Graphical representations of Bayesian cluster analyses withlocation as a prior, based on data from 15 microsatellite loci. The three plots show population assignment results for different values of K: (a) K = 2; (b) K = 3; (c) K = 4. Each vertical line represents an individual and its assignment to a particular cluster; solid lines separate the 7 sampled locations.

Figure S3.Correlation between geographic distance (km) and genetic differentiation (FST) for yellowtail specimens collected in the northern hemisphere (R = 0.9912), with significance determined by one-sided Mantel tests (p = 0.0821).

Figure S4. Correlation between geographic distance (km) and genetic differentiation (FST) for yellowtail specimens collected in the southern hemisphere (R = 0.5051), with significance determined by one-sided Mantel tests (p = 0.3356).

Figure S5. Ten runs of the program BAYESASS for the seven populations in this study. Each graph represents one population and the proportion fononmigrants estimated for that population for each of the replicate runs.