Plants-pollinator interactions have long been a central focus of biology, yet they are often treated as isolated despite increasing recognition of the importance of indirect interactions. By altering pollinator abundance and behavior, predators of pollinators may indirectly alter plant pollination, reproduction, and selection on floral traits. Here we examine the indirect effects of predatory crab spiders on Calochortus luteus.
We addressed three questions: 1. How do crab spiders affect plant-pollinator interactions? We observed visitation to paired flowers randomly assigned to crab spider present and absent experimental treatments in the field and recorded the frequency and duration of visits, visitor identity, and stigma/anther contact. 2.How do crab spiders affect seed set? We compared the number of seeds produced by wild plants which naturally harbored crab spiders and those that were crab spider free. 3. How do crab spiders affect the strength and direct of phenotypic selection on floral size? We measured flower height, flower width, and seed set of wild plants with crab spiders naturally present, and of randomly-selected, paired control plants with crab spiders naturally absent. Using these data we estimated phenotypic selection gradients on flower traits.
Results/Conclusions
We found that crab spiders significantly reduce both the frequency and duration of insect visits to flowers. However, surprisingly the number of seeds produced by plants which harbor crab spiders is significantly higher than the number of seeds produced by crab spider-free plants (means = 43 and 31 seeds respectively, p = 0.02). While there are several potential mechanisms behind this pattern, we hypothesize that crab spiders could mediate self-pollination of C. luteus. This hypothesis suggests that crab spiders could affect genetic variation and inbreeding depression within plant populations – a prediction which our current work is testing.
We found that flowers utilized by crab spiders as hunting sites were 10% larger than randomly selected control flowers. Further, when crab spiders are absent, plants experienced significant selection for large flowers (S = 0.071, p < 0.01), but when crab spiders were present there was no significant selection on flower size (S = -0.018, NS). These results indicate that when a crab spider is present, flower size does not impact seed set. Again, this could potentially be explained by the hypothesis that crab spider mediated self-pollination. Our findings indicate that pollinator predators likely have significant indirect effects on plant ecology and evolution.