Geographically widespread plant species likely experience abiotic and biotic environmental variation across their ranges.  We assessed how herbivory relates to variation in environment among ten populations of Lindera benzoin, a native shrub of the eastern deciduous forest.  Herbivory was measured as the percent of total leaves with any leaf area missing on 40 plants in each population.  Previous research on this species has shown that light environment influences herbivory.  We measured light as the percent of photosynthetically active radiation transmitted through the forest canopy (%PAR).  The main herbivore of L. benzoin is Epimecis hortaria (Lepidoptera), whose adults frequently deposit their eggs on Liriodendron tulipifera in the canopy and the larvae then migrate to L. benzoin in the understory.  We therefore measured basal area (BA) of L. tulipifera in each population using the wandering-quarter method.  We used regression analysis to assess how herbivory is influenced by %PAR, BA of L. tulipifera and geographic data (latitude, longitude and elevation); all variables were standardized to have a mean of zero and standard deviation of one.  Forty individuals from each population were also transplanted to replicate common gardens; this approach will allow us to evaluate variation in defense or tolerance to herbivory among populations. Results/Conclusions

Overall, we observed both geographic and environmental variation in patterns of herbivory.  There was a statistical trend for herbivory to decrease as %PAR increased; previous studies have shown that L. benzoin plants in gaps and edges have reduced herbivory relative to plants in the shade.  Neither elevation nor longitude affected observed herbivory.  BA of L. tulipifera and latitude significantly affected the percent of leaves with herbivory in a population.  A one unit increase in relative BA of L. tulipifera increased herbivory by 30.2%; this relationship may arise as E. hortaria larvae move from the canopy by ‘silking’ to the L. benzoin plants below.  A one unit increase in latitude decreased herbivory by 44.4%.  Similar latitudinal gradients in herbivory have been observed in other species, but over much larger distances than our populations, which ranged from 39.84ºN to 41.13ºN.  The differences we observed may have led to differentiation among populations, as fitness is reduced by herbivory in L. benzoin. We anticipate that results from the common garden will assess differentiation among populations in terms of herbivory-related traits.