Within-colony relatedness and comparison with the literature
At our sites, within-mound relatedness was significantly higher in monogyne mounds (mean and standard errors: 0.644 ± 0.024) than in polygyne mounds (0.269 ± 0.037; F1,60 = 75.832, P<0.001). However, relatedness in polygyne mounds at our sites was significantly greater than zero (t33 = 7.249,P < 0.001; 95% CI: 0.193-0.344). Relatedness in monogyne mounds was significantly lower than 0.75 (t35 = -4.368, P < 0.001; 95% CI: 0.595-0.693), consistent with our genetic results showing that some colonies (18/38) were headed by a multiply mated queen (Appendix S7), as has been reported previously (Fritz, Vander Meer, & Preston, 2006; Lawson, Vander Meer, & Shoemaker, 2012). Within-mound relatedness did not vary by site (F5,60 = 1.781, P = 0.130), nor did the within-mound relatedness of each social form vary by site (F3,60 = 0.382, P = 0.766).
Based on our analysis of within-colony relatedness coefficients between workers reported in the literature, no other study has analyzed within-colony relatedness between fire ant workers in Texas, USA (Fig 5). Of the five studies conducted in Georgia, USA, four reported coefficients in polygyne populations that were not significantly greater than zero, and one reported a coefficient that was significantly greater than zero (mean: 0.16; Ross et al. 1993; Fig 5). Studies conducted on introduced polygyne populations in Australia and Taiwan were greater than zero (Richlands, Australia: 0.246; Chiayi, Taiwan: 0.1444; Taoyuan, Taiwan: 0.1122). Likewise, coefficients of polygyne fire ants in their native range of Argentina were also greater than zero (Corrientes: 0.24; Formosa: 0.15).