ece3569-sup-0001-FigS1.tifimage/tif115KFigure S1. Plot of Log probability [L(K)] as a function of K averaged over five independent runs of Structure. The Y-error bars are standard deviation and K is the assumed number of genetic clusters. Each plot represents a sample set: (A) sample set A (European wildcats, domestic cats and known hybrids); (B) sample set B (Sardinian wildcats, domestic cats and known hybrids); (C) sample set C (domestic cats, European and Sardinian wildcats); (D) sample set D (European wildcats only). (E) Inference of the number of genetic clusters in the study area: posterior distribution of the number of populations estimated using Geneland.
ece3569-sup-0002-FigS2.tifimage/tif300KFigure S2. Maps of posterior probability of European wildcats identified by Geneland. Samples are split into five clusters (= 5): (1) eastern Alps; (2) Maremma; (3) central peninsular Apennines; (4) southern peninsular Apennines; and (5) Sicily.

Table S1. Description of 35 microsatellite loci used to genotype the cat samples (Felis silvestris) analyzed in this study. Locus identifications (ID) and chromosome assignments are from Menotti-Raymond et al. (2003a,b); the asterisk indicates imperfect dinucleotide microsatellites showing some intermediate alleles; primer tails were labelled to fit the design of eight multiplex sets. The allelic range (in base-pairs) and the observed number of alleles at each locus (NA) are reported. FIS = Weir and Cockerham (1984)'s fixation index for each locus in cat subspecies (data set E) and European wildcat populations (data set F) were computed excluding the admixed genotypes (see 'Results'); *** significant departures from Hardy-Weinberg equilibrium at < 0.001 (< 0.00,028 after Bonferroni-correction for 35 independent comparisons). FIS values at X-linked loci FCA240 and FCA651 were computed only in females (Nf, number of females in samples). M, monomorphic loci in wildcats sampled in Sardinia and Sicily.

Table S2. Identification of the number of K clusters in Structure analyses of cat samples (see Table 2). Optimal K values (in bold) were identified by the maximum increase (ΔK) of the mean Ln posterior probability (Mean lnPD) and of the mean FST values (ΔFST) between subsequent analyses. NA, not analysed.

Table S3. Average proportion membership (Qi) of wildcat populations obtained by Structure with = 2–4 and the “localities as prior model”, using: the three cat subspecies (sample set C); European and Sardinian wildcats (domestic cats excluded); European wildcats (Sardinian wildcats excluded), sampled in eastern Alps, Maremma, central and southern Apennines and Sicily (sample set D, see Fig. 2A)

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.