2990 Ecology, 86(11), 2005, pp. 2990–2997  2005 by the Ecological Society of America INSECT HERBIVORES DRIVE IMPORTANT INDIRECT EFFECTS OF EXOTIC PLANTS ON NATIVE COMMUNITIES JENNIFER A. LAU 1 AND SHARON Y. STRAUSS Section of Evolution and Ecology, 2320 Storer Hall, University of California, Davis, California 95616 USA Abstract. Exotic plant species can affect native plant species both directly via com- petition and indirectly by modifying native species’ interactions with other organisms in the community. Both direct and indirect effects can have strong fitness impacts on the native species and can result in cascading effects throughout the invaded community. Many exotic plant species escape from the herbivores with which they co-evolved, but in our study system, the exotic Egyptian alfalfa weevil (Hypera brunneipennis) has been intro- duced to California and is able to feed both on the exotic plant Medicago polymorpha and on the California native Lotus wrangelianus. Because these two plant species share this common herbivore, the possibility of strong herbivore-mediated indirect interactions be- tween these introduced and native plant species is plausible. We used observational studies and manipulative field experiments to determine whether exotic Medicago affects herbivory on the native Lotus and whether Medicago reduces the fitness of Lotus through direct competitive effects and/or herbivore-mediated indirect ef- fects. Both observational and experimental data indicate that Medicago increases weevil herbivory on the native Lotus. Additionally, Medicago reduced Lotus fitness through direct as well as indirect mechanisms. In 2002, Medicago reduced Lotus fitness even when her- bivores were experimentally reduced, suggesting direct competitive effects of Medicago on the native Lotus. In contrast, in 2003, Medicago reduced Lotus fitness only in the presence of herbivores, indicating that the fitness effects in that year were largely indirect and mediated by herbivores. Our results demonstrate that the net fitness consequences of indirect and direct effects of exotic plant species on native plants can vary temporally, depending on the abundance of members of the herbivore community. Also, rather than limiting the success of exotics, introduced herbivores that feed on exotic plants may provide additional, indirect avenues through which exotic plants can reduce the fitness of native species. Key words: apparent competition; associational susceptibility; biological invasion; Hypera brun- neipennis; indirect effect; Lotus wrangelianus; Medicago polymorpha; plant–insect interaction. INTRODUCTION While the direct effect of exotic plants on native vegetation via competition has been well documented (e.g., Walker and Vitousek 1991, Dunbar and Facelli 1999), less obvious indirect effects of exotics on native species may go undetected (but see Callaway et al. 1999, Grosholz et al. 2000, Brown and Mitchell 2001, Chittka and Schurkens 2001, Irwin et al. 2001, Brown et al. 2002, Suarez and Case 2002, Callaway et al. 2004, Rand and Louda 2004). Indirect effects are strictly de- fined as the effects of one species on another that are mediated by a third species (Wootton 1994). Important indirect effects of exotic plants on native plant species could be mediated by herbivores (Callaway et al. 1999, Sessions and Kelly 2002, Rand and Louda 2004), pol- linators (Brown and Mitchell 2001, Chittka and Schur- kens 2001, Brown et al. 2002), mychorrhizal fungi (Marler et al. 1999), endophytic fungi (Clay and Holah Manuscript received 30 November 2004; revised 4 May 2005; accepted 11 May 2005. Corresponding Editor: A. R. Zangerl. 1 Present address: Department of Plant Biology, University of Minnesota, 1445 Gortner Avenue, St. Paul, Minnesota 55108 USA. E-mail: lauxx087@umn.edu 1999, Clay 2001), or parasites (Irwin et al. 2001). Such indirect effects can take many forms (e.g., indirect mu- tualism, apparent competition) and can influence pop- ulation growth (Power 1990, Wootton 1993), com- munity structure, and species coexistence (Holt 1977, Menge 1995, Schmitz 1998, 2003). While some indi- rect effects are predictable functions of the products of direct effects among species, other indirect effects can only be detected experimentally and are not inherently predictable (Wooton 1994). Thus, if exotic species commonly exert strong indirect effects, the community consequences of biological invasions may be difficult to predict and detect. If exotic herbivores invade with exotic plants or if exotic plants are fed upon by generalist herbivores al- ready present in the community, then exotic plants may indirectly affect co-occurring native plant species via these natural enemies (Holt and Lawton 1994, Louda et al. 1997, Rand and Louda 2004). Similar indirect effects mediated by shared predators have been doc- umented between native and exotic avifauna and a shared disease (van Riper et al. 1986) and between exotic and native phytophagous insects and shared par-