Asymmetric competition between plant species.

Despite extensive interest in the role of plant size in competition, few formal attempts have been made to quantify the magnitude of asymmetric competition, particularly for interactions between members of different species. This paper introduces the concept of asymmetric interspecific competition at the population livel (i.e. mean plant performance) in mixtures of species. It proposes an index of interspecific competitive asymmetry which allows for a progressively greater asymmetric effect as the average size differences between competing species increase, and allows for such an effect whether individuals of focal species are larger or smaller, on average, than competitors. This index of competitive asymmetry is evaluated in the study of interactions between two widely coexisting annuals of disturbed habitats, Stellaria media and Poa annua. An experiment was conducted in which the density, relative frequency and relative seedling sizes (emergence times) of Poa and Stellaria individuals were varied. The relative growth rate (RGR) for both species was measured over a 22-day period. An inverse linear model was fitted for each species, relating the RGR of the focal species to the initial biomass of each species. Each response model included an asymmetry coefficient (β) to assess whether the impact of a unit of initial biomass of the associate species changed with the relative sizes of seedlings of the two species. A zero value of β implies symmetric competition between the two populations; i.e. the competitive effect of a unit of associate species biomass does not change with its initial seedling size. If β is positive the smaller the initial relative size of seedlings of the associate species, the smaller their per unit biomass effect on the response of the focal species. The model fitted our data for Stellaria and Poa well and was validated by an alternative modelling approach. Asymmetry coefficients were estimated as 0.508 (P<0.05) for the effect of Poa in the Stellaria model, and 0.0001 (NS) for the effect of Stellaria in the Poa model; i.e. the effect of Poa on Stellaria was asymmetric while the effect of Stellaria on Poa was symmetric. Differences in interspecific species asymmetric competitive effects are discussed within the context of shoot architecture, and the relative importance of competition for light versus soil resources. Finally, we discuss the relationship of this model to earlier models of competitive asymmetry, and consider the implications of interspecific competitive asymmetry for a number of current theories of plant competition and community organisation.