DIPLOID MALES AND COLONY-LEVEL SELECTION IN FORMICA ANTS

It is suggested that the evolution of polyandry by social hymenopteran queens is caused by colony-level selection, either because polyandry affects the distribution of non-functional diploid males in colonies (the load hypothesis) or because it increases the genetic diversity of the worker force (the diversity hypothesis). Diploid males that arise from fertilized eggs that are homozygous at the sex-determining locus (or loci) are inviable or infertile. Models of the load hypothesis analysed in this study suggest that slow growth and high mortality of colonies with diploid males favour single mating by queens. The longer the period of colony growth (the period with selective differences) and the heavier the mortality, the stronger is the selection for monandry. The load hypothesis also predicts an association between monogyny and monandry. In contrast, the diversity hypothesis predicts an association between monogyny and polyandry, as multiple mating offers a way by which a monogynous colony could increase its genetic heterogeneity. Up to 10% of all males are diploid in species and populations of Formica ants with highly polygynous colonies (F. aquilonia, F. polyctena, F. truncorum). No diploid males were found in two mainly monogynous species (F. exsecta, F. pratensis) which also have a high level of monandry. This agrees with the prediction of the load hypothesis. A surprisingly high frequency of nests in three other species (F. rufa, F. lugubris, F. truncorum) with monogynous/weakly polygynous colonies produce diploid males, although the frequency varies among their populations. In extreme cases half of the diploid sexuals within a colony develop into males. Diploid males have been observed only at the time of normal sexual production. It seems that at other times they are eliminated at early developmental stages, so as to minimize the load on the colony.