Correlated Changes in Plant Size and Number in Plant Populations

The capacity of plants to react to the competitive stress of their neighbours by means of phenotypic plasticity has been shown in a large number of experimental investigations. However, the commonest fate of plants under natural conditions is to die before reaching reproductive maturity; this is a basic tenet of Darwinian natural selection theory, but one which has not received much attention, beyond the casual observation that not all seedlings survive to reproduce themselves. Harper this mortality was greatest at the higher density and in the fertilized soil. Several passing references are made by various authors to mortality in experimental plots, but the first formal analysis of the thinning process appeared in a series of examples cited by Yoda et al. (1963). They examined thinning in overcrowded pure populations of several species on soils of varying fertility and found that the number of surviving plants could be related to their mean weight as w = C.p-3/2, where p is the density of surviving plants and w their corresponding mean weight; the value of C varies with the species. The gradient of the line relating p to w remains constant with time while thinning proceeds. They concluded from observations made under natural and cultivated conditions that 'whatever the factors responsible for plant size may be, age, edaphic or other habitat conditions, the possible maximum density for plants of a given average size w, brought about by selfthinning, is expected to be equal to (C/w)2/3. If the age of stands were uniform, stands on soils of different fertilities would be arranged along the line having the gradient value of -3/2 on log/log co-ordinates, those on the more fertile soil exhibiting lower plant density and larger plant size.' In this paper the above generalization is extended to several other species, using data derived from experiments under cultivated conditions and from forestry management tables. Firstly an experiment specifically designed to observe thinning is reported in some detail. This is followed by analyses of data from other published studies which contain sufficient relevant information to construct the necessary density/mean dry weight graphs. As a preliminary example the archetype of these experimental studies is shown