| This is a study of the intra- and inter-specific competitive interactions of the C;The first two chapters each present a set of regression-based empirical models of plant competition, constructed from measurements of growth and water uptake of greenhouse-grown individual plants. The models were constructed from a set of plants ('model' plants) grown without competition under several watering regimes. Each model was then tested on 'test' plants grown, singly or in pairs, under watering regimes different than those of the 'model' plants. Predictions of mean final biomass by the best performing models deviated less than 7% from the true values. Models that used only plant size were outperformed by models explicitly including water as a limiting resource, in agreement with the expectation that a more 'mechanistic' approach will give better predictions.;Ratibida columnifera had a higher relative growth rate but lower water use efficiency (WUE) than Bothriochloa ischaemum (Chapter 3). In both species WUE was highest in young plants and declined with age. In R. columnifera, both a greater WUE and a greater relative allocation to below-ground tissues was associated with lower soil water contents. No such responses to soil water content were found in B. ischaemum. |