COS 88-8 - Implications of a novel mutualist for fecundity metrics of two endemic Hawaiian plants

Wednesday, August 8, 2012: 10:30 AM
Portland Blrm 254, Oregon Convention Center
Austin B. Aslan, Tropical Conservation Biology and Environmental Sciences, University of Hawaii - Hilo, Hilo, HI and Patrick J. Hart, Biology, University of Hawaii at Hilo, Hilo, HI
Background/Question/Methods

A growing literature discusses the importance of non-native species as mutualists with native species. If native mutualists have become extinct, non-natives able to carry out the same functions may act as substitutes. However, non-native mutualists may differ from the extinct natives, either behaviorally or morphologically. For native partner species, such differences may hold ramifications that are hard to predict based solely on interaction frequency. In Hawaii, the non-native Japanese white-eye (Zosterops japonicus) now pollinates the endemic species Clermontia parviflora and C. montis-loa, both of which were formerly pollinated by Hawaiian honeycreepers. The goal of this study was to assess the likelihood that the novel pollinator in this system will impact plant reproductive success by altering the spatial dynamics of pollen flow. We used radio telemetry to document individual Z. japonicusmovements during the plants’ flowering season, estimating 24-hour movement patterns using minimum density kernels. To assess the importance of spatial distance of pollen transfer to plant reproductive components, we performed manual pollination of flowers with pollen from five distance categories, ranging from selfed to 20 km. Resulting seeds were evaluated for viability and percent germination using split-plot ANOVA with germination rate, percent viability, or fruit set as the response variable and site, individual plant, and treatment as explanatory variables.

Results/Conclusions

Telemetry results indicated high local fidelity of birds, which demonstrated largely intrapatch movements. Meanwhile, the split-plot ANOVA revealed that the germination rate of the intermediate-distance pollination crosses performed significantly better than the germination rates of the other four categories of distance for the C. parviflora trial, (Df=29; t=2.456; P=0.0203). However, preliminary results indicate no significant differences among treatments for C. montis-loa. The results for C. parviflora suggest strongest offspring performance at intermediate distances of pollen transfer, supporting the hypothesis that these species are spatially structured. Because white-eye movement patterns appear to support pollen transfer within patches, the intermediate boost in fitness for the C. parviflora pollination regime could carry significant ramifications for plant populations over the long term.