Impact of flower-dwelling crab spiders on plant-pollinator mutualisms

Indirect effects in interactions occur when a species influences a third species by modifying the behaviour of a second one. It has been suggested that indirect effects of crab spiders (Thomisidae) on pollinator behaviour can cascade down the food web and negatively affect plant fitness. However, it is poorly understood how different pollinator groups react to crab spiders and, thus, when a reduction in plant fitness is likely to occur. Using continuous video surveillance, we recorded the behaviour of pollinators on two flower species and the pollinators’ responses to three crab spider treatments: inflorescences (1) with a pinned dried spider, (2) with a spider model made of paper, and (3) without spiders (control). We found that pollinators avoided inflorescences with dried spiders only on one plant species (Anthemis tinctoria). Pollinators showed no significant avoidance of paper spiders. Honeybees and bumblebees did not react to dried spiders, but solitary bees and syrphid flies showed a strong avoidance. Finally, we found no evidence that inflorescences with dried spiders suffered from a decrease in fitness in terms of a reduced seed set. We hypothesise that top-down effects of predators on plants via pollinators depend on the degree of specialisation of pollinators and their tendency to avoid spiders.