Do competing honey bees matter? Dynamics and abundance of native bees before and after honey bee invasion

To provide replicate samples of local bee populations in a nature preserve, Light traps operated continuously on Barro Colorado Island (BCI), Panama, collected bees for 17 years, including 10 years following invasion by Afric an Apis mellifera. Honey bees appeared in light traps as the first swarms c olonized the Panama Canal area. Their numbers followed seasonal trends show n in independent studies, thus indicating bee abundance and activity in a l arge area. No measurable population-level impact of competition between thi s invading honey bee and native bees, despite many demonstrations of resour ce competition at flower patch and colony levels, changed annual abundances of all 15 native bee species. Native bee abundance did not decrease, nor d id native bees show substantial reciprocal yearly change with honey bee abu ndance. One strong negative correlation of bee catches with an extremely ra iny year was found. However, multiple regression using rainfall and honey b ee abundance as the independent variables showed that neither was responsib le for bee population change over 17 years. Nearly half the native species declined during a year that displayed peak honey bee number. That competiti on from honey bees on an island the size of BCI was necessarily reduced bel ow impact levels expected on the mainland is discussed using a model of res ource and consumer density, foraging range, and island size.