THE ROLE OF GROWTH FORM AND CORRELATED TRAITS IN COMPETITIVE RANKING OF 6 PERENNIAL RUDERAL PLANT-SPECIES GROWN IN UNBALANCED MIXTURES

The competitive abilities of six perennial ruderal plants of three dif ferent growth forms were compared via yield measures using an additive diallel experimental design with unbalanced mixtures (9:3 or 3:9 plan ts per pot. respectively). Thus, in a given mixture species A was grow n in two configurations: three individuals in centre position of the p ot together with nine plants of species B in border position and vice versa. Effect competitive abilities as well as response competitive ab ilities of the species were significantly related to canopy height and plant biomass. The species with lower rosette growth form and smaller biomasses were weaker competitors than the species possessing elevate d canopies along with higher biomasses, whereas total leaf area was no t significantly correlated with competitive ability between species. S pecies differences in competitive ability were stronger between the pl ants grown in the central position than between those grown in the bor der position. Furthermore, interactions between species-specific trait s and configuration could be observed. indicating the importance of sp ecies proportions and arrangement patterns for evaluation of competiti ve outcome in the field. The degree of complete transitivity of the co mpetitive network of the six ruderal species, which was significantly higher than expected under the null model in our experimental design, also seemed to depend on species proportions in mixture. Shifts in roo t:shoot ratio of the centre plants when faced with competition by the bolder plants were in the direction of higher shoot allocation for the weak; competitors with rosette growth form irrespective of the neighb our species. except for Bunias orientalis, which showed a more plastic response. The stronger competitors showed higher root allocation (Urt ica dioica) or were hardly affected at all. Consistent with the result s of our experiment, the weaker competitors occur at rather frequently disturbed and therefore transient field sites whereas the stronger co mpetitors predominate at infrequently disturbed sires during several p hases of succession. (C) Elsevier, Paris.