Exploitative competition alters bee foraging and flower choice

In this field experiment, we test and support the hypothesis that exploitative competition between bees can influence several aspects of their foraging behaviour. Three treatments of lavender patches were set out: bumble bees excluded, honey bees excluded, control. Bumble bees are known to handle lavender flowers more rapidly than honey bees, partly due to their longer tongues. As predicted, excluding these superior competitors consistently (n = 4 trials) and greatly increased honey bee numbers per patch (14-fold increase; P < 0.001). The exclusion of bumble bee also caused multiple changes to honey bee foraging behaviour: time spent on a patch (+857 %; P < 0.001), flower handling time (+16 %, P = 0.040), interval between probed flowers (-27 %, P = 0.012), proportion of interflower flights (-26 %, P < 0.001) and flowers rejected (-12 %, P < 0.001). Conversely, and also as predicted, excluding honey bees had no effect on bumble bee numbers or foraging behaviour. A key consequence of bumble bee exclusion was to increase the mean flower nectar content from 0.007 to 0.019 mu l (+171 %). By constructing an energy budget, we find that this leads to honey bees making a substantial, rather than a marginal, energetic profit per flower visited. Our results show the foraging behaviour of individual bees is extremely flexible and greatly influenced by the effects of interspecific competition on nectar rewards. Collectively, these individual decisions can have rapid and important consequences at the community level, including competitive exclusion.