Size and asymmetry: are there costs to winning the royalty race?

Genotypes and phenotypes in a Wolbachia-ant symbiosis

The fitness effects of overt parasites, and host resistance to them, are well documented. Most symbionts, however, are more covert and their interactions with their hosts are less well understood. Wolbachia, an intracellular symbiont of insects, is particularly interesting because it is thought to be unaffected by the host immune response and to have fitness effects mostly focussed on sex ratio manipulation. Here, we use quantitative PCR to investigate whether host genotype affects Wolbachia infection density in the leaf-cutting ant Acromyrmex echinatior, and whether Wolbachia infection density may affect host morphology or caste determination. We found significant differences between host colonies in the density of Wolbachia infections, and also smaller intracolonial differences in infection density between host patrilines. However, the density of Wolbachia infections did not appear to affect the morphology of adult queens or likelihood of ants developing as queens. The results suggest that both host genotype and environment influence the host-Wolbachia relationship, but that Wolbachia infections carry little or no physiological effect on the development of larvae in this system.

The evolution of manipulative cheating

A social cheat is typically assumed to be an individual that does not perform a cooperative behaviour, or performs less of it, but can still exploit the cooperative behaviour of others. However, empirical data suggests that cheating can be more subtle, involving evolutionary arms races over the ability to both exploit and resist exploitation. These complications have not been captured by evolutionary theory, which lags behind empirical studies in this area. We bridge this gap with a mixture of game-theoretical models and individual-based simulations, examining what conditions favour more elaborate patterns of cheating. We found that as well as adjusting their own behaviour, individuals can be selected to manipulate the behaviour of others, which we term 'manipulative cheating'. Further, we found that manipulative cheating can lead to dynamic oscillations (arms races), between selfishness, manipulation, and suppression of manipulation. Our results can help explain both variation in the level of cheating, and genetic variation in the extent to which individuals can be exploited by cheats.

Sperm mixing in the polyandrous leaf-cutting ant Acromyrmex echinatior

The insemination of queens by sperm from multiple males (polyandry) has evolved in a number of eusocial insect lineages despite the likely costs of the behavior. The selective advantages in terms of colony fitness must therefore also be significant and there is now good evidence that polyandry increases genetic variation among workers, thereby improving the efficiency of division of labor, resistance against disease, and diluting the impact of genetically incompatible matings. However, these advantages will only be maximized if the sperm of initially discrete ejaculates are mixed when stored in queen spermathecae and used for egg fertilization in a "fair raffle." Remarkably, however, very few studies have addressed the level of sperm mixing in social insects. Here we analyzed sperm use over time in the highly polyandrous leaf-cutting ant Acromyrmex echinatior. We genotyped cohorts of workers produced either 2 months apart or up to over a year apart, and batches of eggs laid up to over 2 years apart, and tested whether fluctuations in patriline distributions deviated from random. We show that the representation of father males in both egg and worker cohorts does not change over time, consistent with obligatorily polyandrous queens maximizing their fitness when workers are as genetically diverse as possible.