No abstract

Using a comparative approach, we tested the prediction that relative competitive performance of plant species is correlated with distribution along natural gradients of fertility and standing crop. The mean position of 40 species of herbaceous plants along standing crop, phosphorus, nitrate, magnesium, potassium pH, and percent organic content gradients was calculated based on data from 217 quadrats from lake and river shorelines in Nova Scotia, Quebec, and Ontario. Competitive performance was measured in an outdoor experiments in which species were grown together with a common phytometer, Lythrum salicaria (n = 5 replicated per species). The relative ability of each species to suppress the growth of the phytometer was used as a measure of relative competitive performance. This measure of competitive performance was significantly correlated with the mean position of species on the standing crop (r = 0.081; P < 0.0001), percent organic content (r = 0.71; P < 0.0001); phosphorous (r = 0.70; P < 0.0001), nitrate (r = 0.67; P< 0.0001), magnesium (r = 0.66; P < 0.0001), and potassium gradients (r = 0.61; P < 0.0001). When monocotyledons and dicotyledons were considered separately, the relationship between distribution along the macronutrient gradients and competitive performance was much stronger for monocotyledons. It is suggested that this may reflect a differential influence of disturbance on distribution.

No abstract

No abstract

Summary 1We tested the hypothesis that competitive hierarchies are invariant with respect to changing nutrient supply. 2 The competitive performance of 26 shoreline plant species was determined experimentally as the relative ability to suppress the growth of a common indicator (phytometer) species, Penthorum sedoides. Each species was grown with the phytometer under each of two nutrient treatments created with di erent concentrations of a modi®ed Hoagland's solution (n 5 replicates per species/treatment), for two growing seasons. 3 Although shifts in ranking of relative competitive performance were apparent between nutrient levels, competitive performance under high and low nutrient conditions was signi®cantly correlated in both year 1 (r 0.65) and year 2 (r 0.76), when all species were considered. 4 At the broad community scale, the outcome of competitive interactions thus appears to be relatively predictable and independent of the environment, and therefore provides a useful tool for exploring and understanding community pattern. These results cannot address questions related to the outcome of competitive interactions between similar species or the e ects of ®ne-scale pattern.