SEARCH

SEARCH BY CITATION

TABLE S1. Primer pairs used to generate the 534 repeatable, polymorphic AFLP loci used in our analyses. The number of loci yielded by each primer pair is also provided.

TABLE S2.Baseline information about population variability. For AFLPs, the mean (s.e.) within-population expected heterozygosity (i.e., Nei's gene diversity within populations) is provided, as estimated in AFLP-SURV following the methods of Lynch and Milligan (1994). For mtDNA, within-population pairwise sequence divergence is provided, as estimated in MEGA v. 3.1 (Kumar et al. 2004) and corrected for multiple hits using the Kimura 2-parameter model (Kimura 1980). Also reported is host-plant patch size estimated from aerial photographs as in Sandoval (1994a) and Nosil et al. (2003). Host-plant patch size is correlated with insect population size (Sandoval 1994a).

TABLE S3. Detailed summary of behavior for all 78 loci detected as outliers at the 95% quantile level. ‘X’s denote particular population pairs in which a locus was observed to be an outlier. Also shown is which loci were outliers in the ‘global analysis’ where individuals were pooled into groups corresponding to the two hosts (see text for details). The detailed information in this table was used to assemble the outlier classification scheme reported in Table 3. The number of each population pair corresponds to the numbered comparisons listed in Table 2 and Figure 2 of the main text.

TABLE S4. Results of simple and partial mantel tests examining how FST, estimated using all AFLP loci within a class, is related to indices of adaptive divergence and to geographic distance. All distance matrices were log-transformed. Results are also shown for mtDNA divergence, where the only significantly positive associations were detected (in bold). The simple mantel test of genetic distance against geographic distance is the same for each index of adaptive divergence. AD = Adaptive Divergence. GeoD = Geographic Distance. Analogous analyses using raw distance matrices are presented in the main text.

TABLE S5. Summary of the distribution of mantel correlation coefficients from analyses of FST at individual loci against adaptive divergence and geographic distance (n =206, of which 155 were non-outliers in the genome scan). Distance matrices were log-transformed prior to analysis (three loci were excluded due to negative distance in all comparisons). Binomial tests examine whether significant associations were positive more often than expected by chance. One-sample t-tests examine whether the mean of a distribution differs from zero (d.f. = 1 minus the sample size). Significant results are in bold. The simple mantel test of genetic distance against geographic distance is the same for each index of adaptive divergence. AD = Adaptive Divergence. GeoD = Geographic Distance. No. sig. + = number of loci exhibiting a significant positive relationships at p < 0.05; in parentheses are the number of loci with significant positive associations that were outliers in the Dfdist analyses. No. sig. - = number of loci exhibiting a significant negative relationship at p < 0.05. Analogous analyses using raw distance matrices are presented in the main text.

Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
EVO_299_sm_TablesS1-S5.doc163KSupporting 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.