COS 94-2 - Ecological sorting and character displacement in communities of pollinator-sharing Clarkia species

Wednesday, August 9, 2017: 8:20 AM
D139, Oregon Convention Center
Katherine E. Eisen and Monica A. Geber, Ecology and Evolutionary Biology, Cornell University, Ithaca, NY
Background/Question/Methods

Niche differences that promote the co-occurrence of closely related species can arise via two processes: ecological sorting and character displacement. While ecological sorting and character displacement were both first articulated as processes driven by resource competition, shared interactions with predators, mutualists, or facilitators can also shape these processes. One type of biotic interaction that can structure communities occurs among plants that share pollinators, as pollinators are essential for reproduction in many plants. We assessed the consequences of species co-occurrence for pollination-related trait expression in communities of Clarkia (Onagraceae) in the southern foothills of the Sierra Nevada (Kern County, CA). Measurements of plant and floral traits were taken on greenhouse-grown progeny of plants from communities that contain one, two or four species of Clarkia. We analyzed these data across species and within species across community types using a linear-mixed models framework to address two questions: (1) Are species that co-flower more similar (habitat filtering) or more different (competitive exclusion) in their traits relative to species that do not co-flower and directly share pollinators? (2) Do species’ traits that contain genetic variation have different values where they co-occur with other Clarkia species relative to where they occur alone (character displacement)?

Results/Conclusions

We observed patterns of trait variation consistent with habitat filtering, competitive exclusion, and character displacement. In comparisons of species trait values, we found that values of floral diameter were clustered within flowering time periods, which is expected under habitat filtering, while style length displayed an overdispersed pattern within flowering time periods expected under competitive exclusion. All floral and most plant traits exhibited genetic variation in all four species, a requirement for character displacement, and a number of pairs of species exhibited patterns of variation in three traits consistent with character displacement. Two pairs of species (C. unguiculata and C. cylindrica; C. unguiculata and C. speciosa) diverged in two floral traits (floral diameter, petal area) in sympatry relative to in allopatry. One pair of species (C. unguiculata and C. cylindrica) converged in the date of first flower in sympatry while another pair of species (C. speciosa and C. xantiana) diverged. These results indicate that pollinators may affect the traits of co-occurring Clarkia species. More broadly, this study highlights the utility of common garden experiments for identifying patterns that suggest the time scales at which species interactions contribute to community structure.