Although social insect colonies are most easily conceptualized as consisting of a single, once–mated queen and her worker progeny, the number of queens per colony and the number of times queens mate varies broadly in ants and other social insects. Various hypotheses have been suggested for the resulting range of breeding systems and social organizations, respectively; one set of hypotheses relating to both queen number and mate number at the same time is a need for genetic variation, especially in relation to disease resistance. We here carry out a comparative analysis using phylogenetic information and, contrary to one non–phylogenetic previous study, we find that polyandry and polygyny are not significantly associated. However, the level of relatedness within colonies, a quantity affected by both polyandry and polygyny, is significantly associated with parasite loads: species with colonies with low relatedness levels have lower parasite loads. Given that, under the variance–reduction principle, selection on queens for mating frequency ought to continue even in polygynous colonies, we suggest that while parasite loads indeed seem to correlate with intra–colony genetic variability, the relationship to polyandry and polygyny may be complex and requires considerably more experimental investigation.