What limits herb biomass in grasslands: competition or herbivory?

Competition and herbivory are two of the main forces shaping plant communities. Although several studies have investigated their impact on plant populations separately, few investigations have examined how they might interact. With the purpose of clarifying the combined roles of competition and herbivory on herb biomass in a grassland, we assessed the effects of different herbivores with reduced grass competition. We conducted a field experiment in 2000-2001 in a British acid grassland (Oak Mead), where we experimentally reduced grass biomass and excluded rabbits, insects and molluscs in a factorial design. Removing the grasses from Oak Mead dramatically increased herb biomass and total above-ground biomass. Herbivore exclusions (i.e. rabbits, insects and molluscs) did not affect total above-ground biomass, but they altered relative abundance of several species. Grass removal and rabbit exclusion had positive interactions on biomass of several herb species, and there were some subtle interactions between different herbivore groups: monocots benefitted when both rabbits and molluscs were excluded, and some herb species had greater biomass when insects and rabbits were absent. We then compared the results with a 10-year experiment that manipulated similar variables in neighbouring grassland (Nash's Field). The comparison between Oak Mead and Nash's Field showed that cessation of herbicide application returns the system to its previous state of grass dominance after 3 years. Therefore, even when herbs were more abundant, they could not prevent reinvasion of the grasses once external factors were removed.