PLANT-GROWTH ANALYSIS - AN EVALUATION OF EXPERIMENTAL-DESIGN AND COMPUTATIONAL METHODS

Various aspects of the experimental design and computational methods u sed in plant growth analysis were investigated. This was done either a nalytically, or by repeatedly simulating harvests from theoretical pop ulations upon which were imposed the underlying growth curves as well as the variability in plant material. In the first part the consequenc es of neglecting an In-transformation of the primary weight data were considered. T-tests are affected in such a way that significant differ ences between treatments show up less readily than in transformed data . A more fundamental point is that most hypotheses on plant weight con cern proportional effects rather than absolute ones. In these cases, a n In-transformation prior to a statistical test is required anyway. Se condly, the accuracy of average RGR estimates was evaluated, Variabili ty in RGR estimation increases linearly with variability in the plant material. It is also strongly dependent on the time interval between h arvests and the number of replicates per harvest. Even with conservati ve values for plant weight variability, the chances of arriving at abe rrant RGR estimates are rather high, Therefore, it is suggested that t he variability in the population is decreased deliberately, unless var iability within the population is itself of biological interest. Third ly, three computational methods to fit dry weight progressions and des cribe time trends in RGR and related growth parameters were evaluated. Although complicated to calculate, the Richards function was superior to polynomials fitted through either the weight data ('polynomial' ap proach), or the classically derived RGR values ('combined' approach).