Search

Mutual policing is thought to be important in conflict suppression at all levels of biological organization. In hymenopteran societies (bees, ants, and wasps), multiple mating by queens favors mutual policing of male production among workers (worker policing). However, worker policing of male production is proving to be more widespread than predicted by relatedness patterns, occurring in societies headed by single-mated queens in which, paradoxically, workers are more related to the workers' sons that they kill than the queen's sons that they spare. Here we develop an inclusive-fitness model to show that a second reproductive conflict, the conflict over sex allocation, can explain the evolution of worker policing contrary to relatedness predictions. Among ants, and probably other social Hymenoptera, workers kill males to favor their more related sisters. Importantly, males are killed at the larval stage, presumably because workers cannot determine the sex of queen-laid eggs. Sex-allocation biasing favors worker policing because policing removes some males (the workers' sons) at low cost at the egg stage rather than at higher cost at the larval stage. Our model reveals an important interaction between two reproductive conflicts in which the presence of one conflict (sex allocation) favors the suppression of the other (male production by workers).\ud \ud

The collection and handling of colony resources such as food, water, and nest-construction material is often divided into subtasks in which the material is passed from one worker to another. This is known as task partitioning. If task; are partitioned with direct transfer of material between foragers and receivers, queueing delays can occur as individuals search or wait for a transfer partner. Changes in environmental conditions and relative number of foragers and receivers affect these delays as well as colony ergonomic efficiency. These delays are used in recruitment in both honeybees and Polybia wasps. This study investigates the distribution of queueing delays and the information content and quality of those delays using a stochastic-simulation model. Information quality increases with colony size. When the relative proportions of foragers and receivers are suboptimal, the group in excess has better information. Individuals can increase information quality of delays by two mechanisms: averaging over consecutive trips and averaging over multiple transfers within a trip where direct transfer occurs. We suggest that multiple transfer occurs in the honeybee in order to improve information quality. \ud

The collection and handling of colony resources such as food, water, and nest construction material is often divided into subtasks in which the material is passed from one worker to another. This is known as task partitioning. When material is transferred directly from one individual to another, queueing delays frequently occur because individuals must sometimes wait for a transfer partner. A stochastic simulation model was written to study the effect of colony size on these delays. Queueing delay decreases roughly exponentially with colony size because stochastic fluctuations in the arrival of individuals are lower in larger colonies. These results support empirical studies of Polybia occidentalis and other theoretical studies of honeybees. The effect of the relative number of individuals in the two subtask groups was also studied. There is a unique optimal ratio of the number of workers associated with each of the subtasks that simultaneously minimizes mean queueing delay and maximizes colony nectar-processing rate. Deviations from this optimal ratio, for example, as a result of forager mortality or changes in nectar productivity that affect foraging trip duration, increase mean queueing delays greatly, especially in smaller colonies.\ud