Does it matter if herbivory is selective? Responses of an endangered herbaceous legume to experimental grazing.

Background:Herbivory and inter-specific competition are major determinants of the relative abundance of plant species in grasslands. Herbivores modify competitive hierarchies among plants through selective foraging and as a result of variation in the relative tolerance of plants to herbivory. Aims:We examined the direct effects of herbivory and the role of herbivore preference on the growth, flowering and abundance of an endangered perennial legume,Cullen parvum(F. Muell.) J.W. Grimes, in a temperate Australian grassland grazed by sheep. Methods:Hand clipping was used to simulate herbivory. In each of two populations, three replicates of the factorial combination ofC. parvumdefoliation (C. parvumgrazed or ungrazed), neighbour defoliation (neighbouring vegetation grazed or ungrazed) and three defoliation frequency treatments were randomly applied to 0.25-m2quadrats containingC. parvumramets. Results:The abundance, growth and flowering ofC. parvumwere similarly reduced by both selective and non-selective herbivory. When neighbouring vegetation but notC. parvumwas defoliated, growth and abundance were similar and in some cases less than when both neighbours andC. parvumwere ungrazed, suggesting little above-ground competitive effect of the grassland sward. Conclusions:The interactions between herbivory and shoot competition by dominant perennial grassy swards had little influence on the growth and abundance ofC. parvum. Rather, direct effects of defoliation dominated and we predict that the species persistence in grazed grasslands will be influenced more by poor tolerance of herbivory than by how herbivores modify the competitive effect of neighbouring vegetation. [ABSTRACT FROM PUBLISHER]