Lilies at the limit: Variation in plant-pollinator interactions across an elevational range

PREMISE OF THE STUDY: Many studies assume climatic factors are paramount in determining species' distributions, however, biotic interactions may also play a role. For example, pollinators may limit species' ranges if floral abundance or floral attractiveness is reduced at range margins, thus causing lower pollinator visitation and reduced reproductive output. METHODS: To test if pollinators influence the a ltitudina I distribution of Erythronium montanum (Liliaceae) at Mount Rainier National Park, we asked whether (1) seed production in this species relies on pollinators, (2) seed production and pollen limitation is greatest at range limits, and (3) pollinator visitation rates (either overall or by individual taxonomic groups) reflect patterns of seed production and pollen limitation. RESULTS: From this three-year study, we established that this plant does rely on pollinators for fruit set and we found that pollen limitation trended toward being higher at the upper range limit in some years, but not consistently year to year. Insect visitation rates did not mirror spatial patterns of pollen limitation, but annually variable pollinator composition suggested differential importance of some pollinator taxonomic groups (specifically, bumblebees may be better pollinators than syrphid flies). CONCLUSIONS: Overall, these results suggest that while pollinators are critical for the reproductive success of this high mountain wildflower, plant -pollinator interactions do not obviously drive the distribution of this species. Nonetheless, high spatio-temporal variability in range -wide plant -pollinator dynamics may complicate responses to climate change.