In order to test whether predictions of mortality and size hierarchy development, as derived from studies of non-clonal plants, are valid for clonal plants, an experiment was carried out with Brachypodium pinnatum and Carex flacca, two rhizomatous herbs, planted at three densities. Competition was studied at two levels of organization, the shoot (ramet) and the clone (the individual plant consisting of all interconnected ramets). Size inequalities (measured as Gini coefficients), size-dependent growth and mortality of both shoots and clones were analysed over a period of two years. The species were grown in monocultures in three treatments that differed in clone density. At the end of both the first and the second year of the experiment there were no significant differences in Gini coefficients of shoot height or weight between treatments. This was true for both species. In the second year, when the number of shoots and the biomass per plot were extremely high and similar for all treatments, size hierarchies (based on height) of shoots born in spring decreased in the course of the summer and height increment was unrelated to shoot height at the beginning of the growth period. There are reasons to assume that the weight of small shoots did not increase in proportion to their height, and hence that the growth of small shoots was suppressed by larger shoots (asymmetric competition). The observation that smaller shoots had a higher mortality risk accords with this conclusion. In the first year of the experiment, clone mortality due to establishment failures was density-dependent. Clones in the largest clone density remained an order of magnitude smaller than those in the smallest clone density, indicating that clone competition was more severe in the former treatment. The Gini coefficients of clone size. based on rhizome dry weight and the number of shoots, did not differ between treatments suggesting that competition between established clones was symmetric. This was also the case in the second year, as shown by the low clone mortality and the fact that rhizome biomass produced in the second year was linearly related to rhizome biomass produced in the first year. Both species gave identical results. We conclude that competition between ramets within clones appears to be asymmetric in the species that we studied, whereas competition between clones is symmetric, even in extremely dense stands where competition is for light. This is fundamentally different from competition in non-clonal plants which is typically asymmetric under such conditions. Reasons for this difference are suggested and some of the implications of the results are discussed.
