Bottom-up changes in the foundation of a trophic system can have resonating effects on an ecosystem, decreasing ecosystem productivity and diversity, and altering predator-prey interactions. A source of harmful bottom-up effects, non-native ornamental plants have historically been promoted by the landscaping industry for their ‘pest-resistant’ qualities and now comprise more than 85% of the 235 naturalized woody invasive species (Reichard and Hamilton 1997; Pimental et al. 2005; Qian and Ricklefs 2006). Previous studies have found that non-native plants, both congeners of native plants and those unrelated to native plants, decrease insect herbivore diversity by 50 and 75%, respectively. The impact on generalist predators, however, is unknown. This study quantified the impact of non-native plants on spiders, a guild of ubiquitous generalist predators. Non-native plants, especially those without common congener native relatives, were predicted to support fewer spiders, fewer species, and less spider biomass than native plants because of the reduction of prey on such plants. Spider communities were sampled on 13 pairs of congeneric native and non-native plants and on 17 native plants and 17 unrelated non-native species in 4 common garden plots.

Results/Conclusions

A consistent trend in the non-congener plant comparisons suggests that non-native plants in this study supported greater spider richness (1.748 +/- 0.233 versus 1.260 +/- 0.118), abundance (3.515 +/- 0.528 versus 2.215 +/- 0.204), and more biomass (2.092mg +/- 0.4mg  versus 1.595mg +/- 0.2 mg) than on native plants. However, the ability of plants to support spider assemblages seems to be genus-specific, as native Acer, Salix Cornus, and Betula consistently supported greater spider richness, abundance, and biomass while non-native Ulmus, Carpinus, Rosa, Viburnum, Tilia, and Rhododendron supported more. In particular, while native Acer supported significantly greater spider abundance than its non-native congener (3 d.f., t=3.77, P=0.033), non-native Ulmus supported significantly greater spider richness and abundance than its native congener (3 d.f., t=3.83, P=0.031; t=3.64, P=0.036). Additionally, non-congener non-native plants supported less spider richness, abundance, and biomass than congener non-native plants, supporting the hypothesis that unrelated non-natives plants can have severe effects on spider communities. Other features, including low plant height, varied leaf densities, and different niches within the spider guild may have affected spider presence on the plants. Results suggest that not all non-native plant species are equally detrimental to spider communities, nor are all native species equally supportive.