Pollinator predators can significantly alter pollinator behavior and abundance yet their effects on plant fitness are rarely considered. Previously we found that predatory crab spiders on the perennial Calochortus luteus (Liliaceae) reduced both pollinator visit frequency and duration by ~60% each. Despite these declines, flowers harboring crab spiders produced significantly more seeds than flowers without crab spiders. Therefore, we hypothesize that crab spiders transfer self-pollen within a flower while hunting. Because C. luteus is self-compatible but lacks a mechanism for autonomous pollination, we predict that crab spiders increase the rate of selfed seed production (inbreeding) and may therefore increase the expression of inbreeding depression within plant populations.
To test this hypothesis, we collected seeds from maternal plants in four treatments: 1) open-pollinated, crab spider present (CS-present); 2) open-pollinated, no crab spider (no-CS); 3) emasculated and outcross-pollinated (outcrossed); and 4) bagged and hand self-pollinated (selfed). We germinated these seeds, grew them in a greenhouse for two years, and measured fitness metrics (e.g. germination, growth rate, dormancy timing, phenology, and mortality). Comparison of offspring fitness metrics among treatments allows us to determine if CS-present offspring have lower fitness - implying greater expression of inbreeding depression - than no-CS offspring.
Results/Conclusions
The fully outcrossed and fully selfed treatments represent two mating system extremes and allow us to interpret the effect of crab spiders on mating systems. Outcrossed offspring have greater survival over two years than selfed offspring (83% v. 67% respectively). No-CS offspring have similar survival to outcross offspring (81%). CS-present offspring have intermediate survival (76%). Fitness metrics of germination, plant size, date out of dormancy, and mortality follow similar trends to the survival data: outcrossed offspring > no-CS offspring > CS-present offspring > selfed offspring. These trends support our hypothesis that crab spiders mediate self-pollination within flowers and consequentially lead to lower fitness, implying greater expression of inbreeding depression. Our results suggest that this population of C. luteus likely has a mixed mating strategy, and that the presence of crab spiders increases rates of selfing. The fitness of long-lived perennials, like C. luteus, which produce flowers after several years of vegetative growth, may be significantly affected by sit-and-wait predators, such as crab spiders, if they directly increase rates of selfing. Small differences in early fitness metrics caused by increased selfing and expression of inbreeding depression could have significant, multiplicative effects on lifetime plant- and population-level fitness.