Both structural complexity of understory vegetation and food availability are known to affect rodent foraging activity. Invasive plants often alter the architectural complexity and food availability of the invaded habitat, which subsequently may modify rodent foraging choices. Because rodents are avid seed consumers, their foraging choices may affect seedling recruitment. Exotic plants may change rodent foraging choices so that they indirectly impact native plants through apparent competition.  Amur honeysuckle (Lonicera maackii, hereafter honeysuckle) is a highly aggressive invasive plant that outcompetes native plant species. We propose that honeysuckle disrupts rodent behavior by adding two components to the native landscape. First, the shrub provides a large food source in the form of fruits. Second, honeysuckle modifies vegetation cover by changing naturally patchy forest understory into dense and nearly continuous vegetation. Our main objective of this study was to determine how these two components affect mice activity and seedling recruitment. We predicted that mice activity would be positively correlated with honeysuckle cover and fruit production, leading to reduced seedling recruitment of natives (i.e. apparent competition). We tested this hypothesis using a fully-crossed factorial design in which honeysuckle fruits and cover were manipulated for two levels each. We used track plates to record mice activity, and mice exclosures within the factorial design to detect the effects of mice on seedling recruitment.
Results/Conclusions

Our results indicate that removal of honeysuckle cover reduced activity of white-footed mice Peromyscus leucopus by 40%, while fruit availability had no effect on P. leucopus activity. This result agrees with the prediction that invasive honeysuckle alters rodent activity by providing cover (i.e. refuge-mediated apparent competition), but not by providing food. The scale chosen for the experiment, plots of 900 m², probably was not large enough to detect the effect of fruits. Results from the rodent exclosure experiment showed that higher activity of mice in honeysuckle infested plots led to a gradient of seedling establishment. Honeysuckle removal areas had on average 128.2 (SE±47.0 individuals for mice excluded cages and 33.2 (SE±17.9) individuals for open access cages. Honeysuckle infested areas had on average 66.8 (SE±31.0) individuals for mice excluded cages and 23.0 (SE±17.3) individuals for open access cages. Although honeysuckle is widely reported to be detrimental to seedling establishment, our findings demonstrate that apparent competition at our study site partially contributes to the detrimental effect of honeysuckle on the diversity and abundance of native seedlings by providing a refuge for native rodents.