Why do honey-bee (Apis mellifera) foragers transfer nectar to several receivers? Information improvement through multiple sampling in a biological system

The task of nectar foraging in honey-bees is partitioned between foragers a nd receivers. Foragers typically transfer a nectar load in the nest as sub- loads to several receivers rather than as a single transfer. Foragers exper ience delays in finding receivers and use these delays to balance the numbe r of foragers and receivers. A short delay results in the forager-recruitin g waggle dance whereas a long delay results in the receiver-recruiting trem ble dance. Several nectar transfers increase the cost of this system by int roducing additional delays in finding extra receivers. We tested four hypot heses to explain the occurrence of multiple transfer. We found no evidence that multiple transfer is due to different crop capacities of foragers and receivers or that it results from extensive trophallactic interactions with nest-mates. Receiver bees frequently evaporate nectar in their mouthparts to hasten the production of honey. The suggestion has been made that multip le transfer is driven by receivers who take partial loads from foragers to enhance nectar evaporation. An alternative suggestion is is that foragers d rive multiple transfer to gain better information on the balance of forager s and receivers. transfer to gain better information on the balance of Mult iple sampling of delay in finding a receiver reduces the standard deviation of the delay mean and so provides foragers with better information than is provided by a single delay. The enhanced-evaporation hypothesis predicts t hat receivers break foragers' first transfer whereas the information improv ement hypothesis predicts foragers break their first transfers. Furthermore , only the information improvement hypothesis predicts a high level of mult iple receptions. Data on transfer break-off and receiver behaviour strongly support the information improvement hypothesis and reject the enhanced-eva poration hypothesis. We suggest that multiple transfer is an adaptive sampl ing mechanism, which improves foragers' information on colony work allocati on, and that multiple sampling is a common feature of social insect societi es.