What makes a good pollinator? Relationship between pollinator traits and pollination effectiveness in apple flowers

Differences in pollination effectiveness (PE) among pollinators have been widely documented. However, the morphological and behavioural traits underlying these differences have been less investigated. We used single-visit pollen deposition to apple flowers to explore the relationship between pollinator traits and PE. Our objectives were to determine which morphological and behavioural traits promote PE and to establish whether effective pollinators shared similar traits. For each flower visit we recorded duration, flower handling behaviour (top-working vs. side-working: flower visited from the top or from the side, respectively) and resource collected (pollen vs. nectar). For each pollinator species we measured body size, tongue length, hair density and hair length. PE depended on flower handling behaviour (top visitors were more effective than side visitors) and was positively correlated with body size and visit duration, but not with hairiness. We found large differences in PE among 11 pollinator species. The mean number of pollen grains deposited per visit ranged from 201.5 (Apis mellifera) to 833.7 (Andrena flavipes). In general, solitary bees (Andrena, Eucera, Osmia) were more effective than social bees (Bombus, Apis). Dipterans and beetles (Eristalis, muscoid flies, Tropinota, Oxythyrea) showed high to intermediate levels of PE. All effective pollinators were top-workers, but otherwise did not necessarily share similar traits. We conclude that there are different ways to be an effective pollinator. Our results provide a mechanism to explain the posi-tive relationship between pollinator diversity and pollination service and can be used to predict PE in bowl-shaped flowers. Our approach can be easily adapted to other flower-pollinator systems.