Behavioural interactions selecting for symmetry and asymmetry in sexual reproductive systems of eusocial species

Lars Witting

Abstract: Understanding the life-history complex of eusociality has remained an enduring problem in evolutionary ecology, partially because natural selection models have considered traits in relative isolation. I aim for a more inclusive model that uses ecological interactions to predict the evolutionary existence of sexual reproduction, sexual reproduction asymmetry, and sex ratios in eusocial species. Using a two-level selection process, with within-population selection on the sex ratio of the sexual caste and between-population selection on the worker sex ratio and the degree of sexual reproduction asymmetry, it is found that a male-haploid genome and a worker caste of pure females is the evolutionary optimum of most initial conditions when, like in eusocial hymenoptera, there is no pair bond between the sexual male and female. That a diploid genome and a worker caste with both males and females is the evolutionary optimum of most initial conditions when, like in eusocial termites, there is a pair bond. That sex-linked genomes may evolve in diploid eusocials, and that the model will not generally maintain sexual reproduction by itself. These results hold for ploidy-levels that behave as quantitative or discrete traits, over a relatively wide range of the relative investment in a sexual male versus sexual female, and for partial sexual systems where the genomic portion with diploid inheritance is either fixed or random.