Abstract
Both size structure and variability (spatial heterogeneity, disturbance, stochasticity, variation in species attributes, etc.) are regarded as regulatory mechanisms of species coexistence. However, none of the models so far proposed consider both size structure and variability simultaneously. A size‐structured variation model for plant‐community dynamics is proposed, which is based on the diffusion model for growth dynamics of plant populations. This model has four functions: (1) mean growth rate of individuals of size x at time t, G(t, x) (species‐specific mean traits, e.g. competitive ability); (2) variance in growth rate of individuals of size x at time t, D(t, x) (stochastic factors due to genetic variation, environmental heterogeneity, spatial variation of individuals, etc.); (3) mortality rate of individuals of size x at time t, M(t, x); and (4) recruitment rate at time t, R(t), as a boundary condition. The interference function for individuals of size x at time t, C(t, x), is introduced, which expresses the degree of interactions between individuals and hence averaged effects of local neighbourhood competition; the G(t, x), D(t, x), M(t, x) and R(t) functions are given in terms of C(t, x). These four functions describe the growth dynamics of individuals of each species in the plant community. Effects of the G(t, x), D(t, x), M(t, x) and R(t) functions on species coexistence in plant communities were evaluated by simulation and the relative importance of the D(t, x) function as well as size structure was shown for species coexistence especially in plant communities where competition among species is non‐transitive or niche limitation does not work.