All organisms face the important question of when to reproduce and how much to invest in offspring. For social insects, such as honey bees (Apis mellifera), reproduction is not the production of individuals but the formation of new colonies. Honey bees can make offspring colonies in two ways: by producing swarms composed of a queen and thousands of workers or by producing male drones. There is a large asymmetry in the cost and expected payoff of producing daughter queens versus drones. Drones are cheap but each one is unlikely to mate successfully, while queens are nearly guaranteed to mate but costly because of the large investment of a swarm. In addition, while drones only contribute directly to colony fitness, the production of workers contributes both directly as part of the colony’s investment in the swarm and indirectly by increasing the colony’s total energy budget. The timing of swarming and drone production are crucial for reproductive success because food resources are seasonal and colonies need to build sufficient stores to survive winter. Using a linear programming model, we ask the question of when a honey bee colony should produce drones and when it should swarm to maximize its total reproductive success.
Results/Conclusions
We find that the optimal behavior for an individual colony is to produce all drones prior to producing swarms, a solution that would be impossible on a population scale because of the need for queens and drones to co-occur. The timing of drone production and swarming are therefore not solely determined by the timing of seasonal resource availability but by the game theoretic problem of coordinating the production of sexual reproductives with other colonies.